AI Post Transformers
By mcgrof
Listen to a podcast, please open Podcast Republic app. Available on Google Play Store and Apple App Store.
Image by mcgrof
Category: Technology
Open in Apple Podcasts
Open RSS feed
Open Website
Rate for this podcast
Subscribers: 0
Reviews: 0
Episodes: 1044
Description
AI-generated podcast where hosts Hal Turing and Dr. Ada Shannon discuss the latest research papers and reports in machine learning, AI systems, and optimization. Featuring honest critical analysis, proper citations, and nerdy humor.
| Episode | Date |
|---|---|
|
Nemotron-Labs-Diffusion: One Model, Three Decoding Modes
|
Nov 07, 2026 |
|
Context Compaction Cuts Costs: "You Don't Need Long Context
|
Nov 07, 2026 |
|
FastKVComp: Checking Whether High-Throughput KV Compression Pays Off
|
Nov 07, 2026 |
|
Cascade: SLO-Aware Latency Budgets for Fair, High-Goodput LLM Serving
|
Nov 06, 2026 |
|
Genuine 1-Bit Post-Training Quantization for LLMs: All for 1-Bit
|
Nov 05, 2026 |
|
PagedMoE: Cross-Precision Expert Caching for SSD-Resident MoE Inference
|
Nov 05, 2026 |
|
Adaptive Mixed-Precision Computing Speeds Up Nearest Neighbor Search
|
Nov 05, 2026 |
|
LeanSparse: Shrinking GPU Memory for Sparse-Attention LLM Serving
|
Nov 04, 2026 |
|
ECHO: Lossless KV Cache Offloading for Sparse Attention LLMs
|
Nov 04, 2026 |
|
Joules-to-GDP: Economic Value Produced Per Unit of AI Energy
|
Nov 04, 2026 |
|
Intelligence per Watt: Measuring Local AI Efficiency Against the Cloud
|
Nov 04, 2026 |
|
How to Loop Your Transformer: Scaling, Specialization, Parallel Inference
|
Nov 03, 2026 |
|
Test-Time Training via Context Distillation: Learning What to Remember
|
Nov 02, 2026 |
|
Speculative Decoding Surprisingly Excels on Sparse MoE Models
|
Nov 01, 2026 |
|
FlatCake: Flat KVCache Storage Beats Tiered Memory Hierarchies
|
Nov 01, 2026 |
|
Agent Memory Costs: Characterizing Write-Time Overhead in Long-Horizon Workloads
|
Nov 01, 2026 |
|
Investor's Last Exam: Benchmarking AI Equity Research Without Future Leakage
|
Nov 01, 2026 |
|
RaBitQ-SSD: Split Codes and Pipelined I/O for SSD Vector Search
|
Nov 01, 2026 |
|
Vector Policy Optimization: Training for Diversity to Improve Test-Time Search
|
Oct 31, 2026 |
|
Harness2: Recursive Self-Improving Agent Harnesses Without Verifiers
|
Oct 31, 2026 |
|
Sliding-Window Attention Beats Linear Attention Without Any Training
|
Oct 31, 2026 |
|
The Illusion of State in State-Space Models
|
Oct 30, 2026 |
|
LLM Evolutionary Search Finds Cheaper Quantum Arithmetic Circuits
|
Oct 30, 2026 |
|
FastPersist: Making Per-Iteration Deep Learning Checkpointing Nearly Free
|
Oct 30, 2026 |
|
Latent Compaction with Test-Time Training: Learning Compact KV Caches
|
Oct 30, 2026 |
|
DepthBench: How Residual Connections Unlock Computational Depth in Transformers
|
Oct 30, 2026 |
|
Gradient Noise Scale Predicts Critical Batch Size for Large-Batch Training
|
Oct 30, 2026 |
|
SanSi: Looped Typed Decision Models for System 1.5 Thinking
|
Oct 28, 2026 |
|
Cartridges++: Compressed KV Memory Without Off-Context Derailment
|
Oct 28, 2026 |
|
Agentic Coding in the Wild: GitHub Copilot's Cache Behavior at Scale
|
Oct 28, 2026 |
|
Hydragen: Shared-Prefix Attention for High-Throughput LLM Inference
|
Oct 27, 2026 |
|
Soup-of-Experts: Pretraining Specialists via Parameter Averaging
|
Oct 27, 2026 |
|
DigitalCoach: Gaps in Human and AI Software Coaching
|
Oct 27, 2026 |
|
Model Soups: Averaging Fine-Tuned Weights Beats Ensembles' Cost
|
Oct 27, 2026 |
|
Joint MoE Scaling Laws: Mixture of Experts Can Be Memory Efficient
|
Oct 27, 2026 |
|
Meta and AMD Co-Design MI350X for Recommendation Model Training
|
Oct 26, 2026 |
|
500xCompressor: Compressing Prompts into a Single Learned Token
|
Oct 26, 2026 |
|
Context Language Models: Letting Models Edit Their Own Context
|
Oct 26, 2026 |
|
HyperTuning via Gisting: Cheap Adapter Generation, Mixed Results
|
Oct 25, 2026 |
|
Formalizing Hopfield Networks and Boltzmann Machines in Lean 4
|
Oct 25, 2026 |
|
How Linear Attention Remembers: Writing and Recalling Facts
|
Oct 25, 2026 |
|
Mixture of a Million Experts: Scaling Routing with Product Keys
|
Oct 25, 2026 |
|
Gist Tokens: Compressing Prompts Up to 26x
|
Oct 25, 2026 |
|
HyperTuning: Adapting Language Models Without Back-propagation Using Hypernetworks
|
Oct 25, 2026 |
|
Scaling Laws for Fine-Grained Mixture of Experts
|
Oct 25, 2026 |
|
Exact Learning Dynamics in Deep Linear Neural Networks
|
Oct 24, 2026 |
|
Compact Proofs of Model Performance via Mechanistic Interpretability
|
Oct 24, 2026 |
|
Formal Lean Proof of PAC Learnability for Decision Stumps
|
Oct 24, 2026 |
|
Backprop as Functor: Composing Learners Through Category Theory
|
Oct 24, 2026 |
|
Verifying Peephole Rewrites in SSA Compiler IRs with Lean
|
Oct 24, 2026 |
|
Multi-Agent Transactive Memory: Sharing Agent Trajectories Across a Population
|
Oct 23, 2026 |
|
DeLM: Decentralized Coding Agents Sharing Context Without a Leader
|
Oct 23, 2026 |
|
MEMORILLA: Latent Semantic Memory Beats Text Retrieval for Frozen LLMs
|
Oct 23, 2026 |
|
PASTA: Steering LLM Attention Toward Highlighted Prompt Text
|
Oct 23, 2026 |
|
Jev-Mem: Fast System-One Controllers for Agentic Memory
|
Oct 21, 2026 |
|
Harness Learning: Training Small Models to Rewrite Code Around Frozen Solvers
|
Oct 21, 2026 |
|
Claude 3.5 Haiku Counts Characters on Curved Feature Manifolds
|
Oct 21, 2026 |
|
Cogentic: Multi-Agent Harness Discovers Proofs for Open Problems
|
Oct 20, 2026 |
|
Chunk-Aware KV Cache Management Bridges LLM Serving and CXL-SSDs
|
Oct 19, 2026 |
|
MAGMA: Multi-Graph Agentic Memory for Long-Horizon AI Agents
|
Oct 19, 2026 |
|
LoCoMo Benchmark: Do LLM Agents Remember Very Long Conversations?
|
Oct 19, 2026 |
|
Emergent Abilities of Large Language Models: Real or Metaphor?
|
Oct 19, 2026 |
|
Measuring Emergent Coordination in Multi-Agent LLMs via Information Decomposition
|
Oct 19, 2026 |
|
Code Predicts, Jev Judges: Separating Simulation from Evaluation
|
Oct 19, 2026 |
|
Memory Caching: Giving Recurrent Models Growing Memory
|
Oct 18, 2026 |
|
Query-Key Normalization Improves Low-Resource Transformer Translation
|
Oct 18, 2026 |
|
BABILong: Testing LLM Long-Context Reasoning in a Haystack
|
Oct 18, 2026 |
|
Raven: Sparse Memory Routing for Long-Range Recall
|
Oct 17, 2026 |
|
MARCH: Content-Routed State Anchors Fix Recurrent Memory Loss
|
Oct 17, 2026 |
|
Automatic Curriculum Learning Through Value Disagreement in Reinforcement Learning
|
Oct 17, 2026 |
|
Emergent Tool Use From Multi-Agent Autocurricula in Hide-and-Seek
|
Oct 17, 2026 |
|
RaReCache: Rank-Disagreement Recomputation for Cross-Model KV Cache Reuse
|
Oct 16, 2026 |
|
KV Cache Translation Across Heterogeneous Large Language Models
|
Oct 16, 2026 |
|
GraniKV: Asymmetric KV-Cache Paging for Shared-Prefix Agents
|
Oct 15, 2026 |
|
Self-Evolving Curriculum: Bandit-Selected Problems for LLM Reasoning RL
|
Oct 15, 2026 |
|
Attention Sinks Enable Streaming Language Models Over Endless Input
|
Oct 14, 2026 |
|
Speculative Prefill: Training-Free Token Importance for Faster TTFT
|
Oct 14, 2026 |
|
In-Context Autoencoder: Compressing Long Contexts into Memory Slots
|
Oct 14, 2026 |
|
LLMLingua: Compressing Prompts with Small-Model Token Pruning
|
Oct 14, 2026 |
|
Raptor: First 3D-DRAM Accelerator for Generative Inference Silicon
|
Oct 13, 2026 |
|
Context Language Models: Letting the Model Rewrite Its Own Context
|
Oct 13, 2026 |
|
Product Keys Give Transformers Million-Slot Memory Layers
|
Oct 13, 2026 |
|
LongAttnComp: Cross-Family Context Compression for Long-Context Reasoning
|
Oct 13, 2026 |
|
AttnComp: Attention-Guided Adaptive Context Compression for RAG
|
Oct 13, 2026 |
|
Jenga: Two-Level Memory Allocation for Heterogeneous LLM Serving
|
Oct 13, 2026 |
|
Memory Layers at Scale: Trainable Key-Value Lookup for Facts
|
Oct 13, 2026 |
|
ATLAS: Full-Stack Simulator for 3D-DRAM LLM Accelerators
|
Oct 13, 2026 |
|
AI R&D Automation and the Intelligence Explosion Question
|
Oct 13, 2026 |
|
MLCommons Chakra: Standardized Execution Traces for AI Benchmarking and Co-design
|
Oct 13, 2026 |
|
Coconut: Reasoning in Continuous Latent Space Instead of Words
|
Oct 12, 2026 |
|
Abstract Chain-of-Thought: Reasoning with Discrete Latent Tokens
|
Oct 12, 2026 |
|
Model-Aware Tokenizer Transfer: Fitting New Vocabularies to Frozen LLMs
|
Oct 12, 2026 |
|
RedFuser: Automatic Fusion of Cascaded Reductions for AI Accelerators
|
Oct 12, 2026 |
|
vLLM Semantic Router: Signal-Driven Routing for Mixture-of-Modality Models
|
Oct 12, 2026 |
|
Workload–Router–Pool: Unifying LLM Routing, Models, and GPU Fleets
|
Oct 12, 2026 |
|
98× Faster LLM Routing with Flash Attention for vLLM Semantic Router
|
Oct 12, 2026 |
|
Alternating Updates: Decoupling Width for Efficient Transformers
|
Oct 10, 2026 |
|
Birkhoff Geometry of Manifold-Constrained Hyper-Connections: Width Fades With Depth
|
Oct 10, 2026 |
|
LayerRoPE: Treating Transformer Depth Growth as a Positional Code
|
Oct 09, 2026 |
|
HBF Sucks? Why Faster Flash Slows KV-Centric LLM Serving
|
Oct 08, 2026 |
|
Synthetic Controls Test Whether AI Drove arXiv Mathematics Surge
|
Oct 07, 2026 |
|
Agensh: Scaling Coding Agent Teams Past a Thousand
|
Oct 07, 2026 |
|
SPLASH: Sparse Attention Co-Designed with High-Bandwidth Flash
|
Oct 06, 2026 |
|
Disaggregated Quantization: Specializing Numeric Formats for LLM Prefill and Decode
|
Oct 05, 2026 |
|
Hot–Cold KV Tiering Across HBM and High Bandwidth Flash
|
Oct 05, 2026 |
|
Fragility of Chain-of-Thought Monitoring Across Diverse Languages
|
Oct 05, 2026 |
|
Attention–FFN Disaggregation Design Space for MoE LLM Serving
|
Oct 04, 2026 |
|
Scaling Multi-Agent LLM Collaboration Past a Thousand Agents
|
Oct 04, 2026 |
|
Small Agents vs. Large Models: A Coordination Test
|
Oct 04, 2026 |
|
Diversity Over Scale: Why Multi-Agent LLM Systems Plateau
|
Oct 04, 2026 |
|
Invisible Reasoning: Filler Tokens Boost LLM Accuracy Unseen
|
Oct 03, 2026 |
|
Sparse Delta Memory: Scaling Linear RNN State Without Extra Compute
|
Oct 03, 2026 |
|
Subliminal Learning: Why Noise Carries Hidden Knowledge
|
Oct 03, 2026 |
|
Subliminal Learning Through Gradient Alignment in MNIST Distillation
|
Oct 03, 2026 |
|
Teaching Claude Why: Fixing Agentic Misalignment Through Safety Training
|
Oct 02, 2026 |
|
HijackKV: Poisoning Position-Independent KV Cache Reuse in RAG Serving
|
Oct 02, 2026 |
|
Reasoning Models Sometimes Output Illegible Chains of Thought
|
Oct 02, 2026 |
|
Stealing Hidden Reasoning Traces from Proprietary LLM APIs
|
Oct 02, 2026 |
|
CacheTrap: Flipping One KV Cache Bit to Hijack LLMs
|
Oct 02, 2026 |
|
Tutti: Making SSD-Backed KV Cache Practical for Long-Context Serving
|
Oct 02, 2026 |
|
KV Cache Offloading for Concurrent Agents: When Does It Help?
|
Oct 01, 2026 |
|
KV-Lingo: Translating KV Caches Between Frozen Models
|
Oct 01, 2026 |
|
RelayS2S: Dual-Path Speculative Generation for Real-Time Dialogue
|
Sep 30, 2026 |
|
Training Transformers for Compressible KV Caches
|
Sep 30, 2026 |
|
JustAskJev: One Question Detects Ten AI Failure Modes
|
Sep 30, 2026 |
|
Gated Convolutions Challenge RNNs in Language Modeling
|
Sep 30, 2026 |
|
More Agents Is All You Need: Sampling Beats Scaling
|
Sep 30, 2026 |
|
BitWhisper: Covert Thermal Signaling Between Air-Gapped Computers
|
Sep 30, 2026 |
|
YOCO: Caching Once for Decoder-Decoder Long-Context Language Models
|
Sep 29, 2026 |
|
Qwen3.8-Next Design: Hybrid Attention, Residuals, and N-gram Embeddings
|
Sep 29, 2026 |
|
mHC: Manifold-Constrained Hyper-Connections Stabilize Wider Residual Streams
|
Sep 29, 2026 |
|
Hyper-Connections: Learning Residual Strengths for Faster Transformer Training
|
Sep 28, 2026 |
|
Subliminal Learning: Hidden Behavioral Traits Transmitted Through Model-Generated Data
|
Sep 27, 2026 |
|
Next-Latent Prediction Lets Transformers Learn Compact World Models
|
Sep 27, 2026 |
|
BASED: Balancing Recall and Throughput in Linear Attention
|
Sep 26, 2026 |
|
ALEX: Building an Updatable Learned Index for Databases
|
Sep 24, 2026 |
|
SALI: Fixing Concurrency Bottlenecks in Learned Indexes
|
Sep 24, 2026 |
|
APEX: A High-Performance Learned Index on Persistent Memory
|
Sep 24, 2026 |
|
Teaching Language Models to Bet Honestly on Their Own Answers
|
Sep 23, 2026 |
|
Structural Language Models of Code: Any-Code Completion with Trees
|
Sep 23, 2026 |
|
Zero-Seed Data Generation: Simula's Reasoning-Driven Pipeline
|
Sep 23, 2026 |
|
Chain-of-Layers: Skipping and Looping Frozen LLM Layers Per Input
|
Sep 23, 2026 |
|
Skip or Loop Layers: Learning Programs-of-Layers in Frozen LLMs
|
Sep 23, 2026 |
|
DeepSeek-V4.1-Flash: Shrinking KV Cache at 552B Scale
|
Sep 23, 2026 |
|
The Unreasonable Ineffectiveness of Deep Transformer Layers
|
Sep 22, 2026 |
|
Transformer Layers as Painters: Skipping and Reordering Frozen LLM Layers
|
Sep 20, 2026 |
|
ShortGPT: Deleting Redundant LLM Layers, Free Lunch or Artifact?
|
Sep 20, 2026 |
|
ProgramBench: Frontier Agents Fail to Rebuild Software from Scratch
|
Sep 19, 2026 |
|
NVQLink: Wiring Supercomputers Directly to Quantum Chips
|
Sep 19, 2026 |
|
The Matthew Effect in RL: Rich Get Richer on Hard Problems
|
Sep 19, 2026 |
|
Full-Bandwidth Transformer: Feeding Hidden States Back Into the Stack
|
Sep 19, 2026 |
|
Redwood: An AI-Designed Accelerator Built and Verified in Two Weeks
|
Sep 19, 2026 |
|
Recursive Self-Improvement and Alignment: OpenAI's "An Alien Mind"
|
Sep 18, 2026 |
|
Cross-Model KV Cache Transfer for Fast LLM Prefill Reuse
|
Sep 17, 2026 |
|
CacheBridge: Fixing Cross-Model KV Cache Transfer Failures
|
Sep 17, 2026 |
|
Latent Space Communication via K-V Cache Alignment
|
Sep 17, 2026 |
|
Mixture-of-Translators: Sharing KV Caches Across Different LLMs
|
Sep 17, 2026 |
|
Dream-RSI: Teaching AI How to Search, Not Just Solve
|
Sep 17, 2026 |
|
Semantic Cache Distillation: Solving Semantic Drift in KV Cache Transfer
|
Sep 15, 2026 |
|
Programmatically Interpretable Reinforcement Learning: Readable Policies
|
Sep 14, 2026 |
|
Deep Drone Acrobatics: Vision-Only Zero-Shot Sim-to-Real Flight
|
Sep 14, 2026 |
|
Imitative Reinforcement Learning for UCAV Pursuit-Lock-Launch Dogfights
|
Sep 14, 2026 |
|
Hierarchical RL Beats an F-16 Instructor Pilot 5-0
|
Sep 13, 2026 |
|
Guided Policy Search: Teaching Neural Nets via Trajectory Optimization
|
Sep 13, 2026 |
|
MPC-Net: Learning Optimal Control via the Hamiltonian
|
Sep 13, 2026 |
|
An Algorithmic Perspective on Imitation Learning
|
Sep 12, 2026 |
|
Reduction of Imitation Learning to No-Regret Online Learning
|
Sep 12, 2026 |
|
Piper: Fixing GPU Utilization in MoE Training on Frontier
|
Sep 11, 2026 |
|
MoM: Mixture-of-Memories Fixes Linear Attention Recall
|
Sep 10, 2026 |
|
Photonic Interconnects and the Race to Cut Prefill Latency
|
Sep 08, 2026 |
|
The Case for Learned Index Structures Revisited
|
Sep 07, 2026 |
|
AI Coaching That Preserves Human Skill Development
|
Sep 06, 2026 |
|
NVIDIA Ultra-CC Beats Top Human at IOI 2026
|
Sep 06, 2026 |
|
The Great Debate: Bounded Refinement vs Open-Ended Self-Improvement
|
Sep 06, 2026 |
|
MegaTrain: Training 120B-Parameter Models on a Single GPU
|
Sep 05, 2026 |
|
Language Models Learn to Fetch Only What They Need
|
Sep 04, 2026 |
|
Disaggregating Prefill, Decode, Attention, and FFN for Agentic Inference
|
Sep 04, 2026 |
|
KV Cache Management Faces Its First Head-to-Head Test
|
Sep 04, 2026 |
|
TreeWY: Speculative Verification for Gated DeltaNet Hybrids
|
Sep 03, 2026 |
|
TimesFM: A Decoder-Only Foundation Model for Zero-Shot Time-Series Forecasting
|
Sep 03, 2026 |
|
Reptile: The First-Order Meta-Learning Shortcut That Works
|
Sep 02, 2026 |
|
FLINT: Sharing High Bandwidth Flash for Scalable LLM Inference
|
Aug 30, 2026 |
|
LinearKV: When Exact State Merging Breaks Hybrid Models
|
Aug 30, 2026 |
|
On-Policy Distillation: Why a Stronger Teacher Can Backfire
|
Aug 28, 2026 |
|
Weak-to-Strong On-Policy Distillation Beats the Teacher
|
Aug 28, 2026 |
|
Token Teachability: Rethinking Disagreement in On-Policy Distillation
|
Aug 28, 2026 |
|
Decoupling KL Direction from Rollout Source in LLM Distillation
|
Aug 28, 2026 |
|
Why AI Systems Don't Learn After Deployment
|
Aug 27, 2026 |
|
Sleeper Memory Poisoning: When Assistants Remember Lies
|
Aug 27, 2026 |
|
Second-Order Optimization Meets Runtime Scheduling at Scale
|
Aug 26, 2026 |
|
Inside Claude Code's Agentic Loop: A Design Space Analysis
|
Aug 25, 2026 |
|
Mechanist: Automating the Discovery of How AI Models Think
|
Aug 25, 2026 |
|
Hand-Written PTX vs WMMA: A Precision-Dependent GPU Speedup
|
Aug 25, 2026 |
|
Grep vs Vector Search: How Agent Harnesses Shape Retrieval Accuracy
|
Aug 23, 2026 |
|
Approaching Shannon Bound: Lossless LLM Weight Compression
|
Aug 22, 2026 |
|
AI-Generated Text and the Death of the Open Web
|
Aug 22, 2026 |
|
Decomposing Speedups Across Runtime, Kernel, and Quantization
|
Aug 22, 2026 |
|
Distributed Shampoo: Making Second-Order Optimization Practical at Scale
|
Aug 21, 2026 |
|
SOAP: Stabilizing Shampoo's Second-Order Optimizer with Adam
|
Aug 21, 2026 |
|
Test-Time Adaptation Through Entropy Minimization
|
Aug 21, 2026 |
|
Preconditioned Optimization Without the Full Matrix Cost
|
Aug 20, 2026 |
|
Scalable Second-Order Optimization: Shampoo at Scale
|
Aug 20, 2026 |
|
Naive Test-Time Adaptation Destabilizes LLM Predictions
|
Aug 16, 2026 |
|
TwinQuant: Manifold-Constrained Low-Rank Decomposition for 4-Bit Quantization
|
Aug 15, 2026 |
|
Volatility Optimization Is Actually Bayesian Inference
|
Aug 15, 2026 |
|
Learning, Fast and Slow: LLMs That Adapt Without Forgetting
|
Aug 15, 2026 |
|
Continual Learning in LLMs: Beyond Catastrophic Forgetting
|
Aug 15, 2026 |
|
Hal:
Memory-Augmented LLM Agents Still Hit Continual Learning's Wall
|
Aug 15, 2026 |
|
In-Place Test-Time Training Turns Fast Weights Into Online Memory
|
Aug 15, 2026 |
|
Model-Agnostic Meta-Learning for Fast Task Adaptation
|
Aug 15, 2026 |
|
Kohonen's 1972 Correlation Matrix Memory, Decades Before Attention
|
Aug 15, 2026 |
|
TFGN: Replay-Free, Task-Free Continual Pre-Training at Scale
|
Aug 12, 2026 |
|
StrataCL: Fabric-Native Zero-Copy Comms for AI Supernodes
|
Aug 11, 2026 |
|
SnapStream: Taming KV-Cache Eviction for Dataflow Accelerators
|
Aug 11, 2026 |
|
Move the Query, Not the Cache: MLA Rewrites GPU Fabric Attention Routing
|
Aug 10, 2026 |
|
MiCA: Mining Minor Singular Directions for Knowledge Injection Beyond LoRA
|
Aug 10, 2026 |
|
Silicon Showdown: GPU vs Apple Silicon LLM Inference Limits
|
Aug 09, 2026 |
|
Global Memory Bloat in Long-Context LLM Serving
|
Aug 04, 2026 |
|
FreeAct: Rethinking One-to-One Transforms for LLM Quantization
|
Aug 04, 2026 |
|
Making Every Verified Token Count in MoE Speculative Decoding
|
Aug 04, 2026 |
|
Model Predictive Control's Real-Time Structure, from Chapter to Cockpit
|
Aug 04, 2026 |
|
DualDecoder: Fixing GPU Memory Bloat in Long-Context KV Cache Offloading
|
Aug 04, 2026 |
|
Distributed Weight Data Parallelism Cuts LLM Inference Stalls
|
Aug 02, 2026 |
|
Adapting Without Forgetting: A Lifelong Learning Roadmap for LLM Agents
|
Aug 02, 2026 |
|
Data Temporality's Hidden Impact on LLM Pretraining
|
Aug 02, 2026 |
|
Test-Time Training Turns EDA Feedback Into Live Weight Updates for RTL
|
Aug 01, 2026 |
|
AdaJEPA: Self-Adapting Latent World Models via Test-Time MPC
|
Aug 01, 2026 |
|
Cross-Family Speculative Prefill Cuts Long-Context Latency
|
Aug 01, 2026 |
|
MemPO: Teaching Agents to Write Their Own Memory
|
Jul 31, 2026 |
|
The Unlearnability Phenomenon in RLVR Reasoning Models
|
Jul 31, 2026 |
|
Main Trust Issue in FPGA HLS Design Workflow
|
Jul 31, 2026 |
|
Peer-Preservation: When Frontier Models Protect Other AIs
|
Jul 30, 2026 |
|
Superhuman Adaptable Intelligence Challenges the Idea of AGI
|
Jul 30, 2026 |
|
Eigenvectors of Experts: Training-free MoE Routing Without Collapse
|
Jul 30, 2026 |
|
Adaptive Block-Scaled Data Types for FP4 Training
|
Jul 30, 2026 |
|
Dual-Path NVMe-Direct KV-Cache Offloading for Edge LLM Inference
|
Jul 29, 2026 |
|
Kimi K3 Goes Trillion-Scale With Sparse MoE Efficiency
|
Jul 28, 2026 |
|
NaturalReasoning: Backtranslating Reasoning Questions at Scale
|
Jul 27, 2026 |
|
Hilbert Operator Reveals What Networks Actually Learn
|
Jul 27, 2026 |
|
Agentic Hardware Design: Evolving RTL Through Repository-Level Code Loops
|
Jul 26, 2026 |
|
MegaScale-Infer: Disaggregating Experts for Faster MoE Serving
|
Jul 26, 2026 |
|
ASAP: Disaggregating Attention and Experts for Faster MoE Prefill
|
Jul 26, 2026 |
|
Multi-Agent AI Rewrites a Million-Line Chip Design Tool
|
Jul 25, 2026 |
|
Small Collectives, Big TLB Cost: Reverse Address Translation in Scale-Up GPU Pods
|
Jul 24, 2026 |
|
Compute or Load? Cake's Smart KV Cache Scheduler
|
Jul 24, 2026 |
|
Fast State Restoration for Evicted LLM KV Caches
|
Jul 24, 2026 |
|
IMPRESS: A Multi-Tier KV Storage System for Faster LLM Prefill
|
Jul 24, 2026 |
|
CacheFlow: Optimal 3D-Parallel KV Cache Restoration
|
Jul 24, 2026 |
|
Context is All You Need: Fixing OOD Drift Without Retraining
|
Jul 23, 2026 |
|
Post-Training Science: Scaling Laws for SFT and LoRA
|
Jul 23, 2026 |
|
Prompt Boundary-Aware Scheduling with Event Tensors for Dynamic Kernels
|
Jul 23, 2026 |
|
The Reversal Curse: When A Is B But Not B Is A
|
Jul 23, 2026 |
|
How Patchscopes Reveals What Language Models Really Think
|
Jul 22, 2026 |
|
Language Model Continual Learning: Do Written Facts Survive Repeated Weight Updates?
|
Jul 22, 2026 |
|
The First Fact That Wouldn't Reason
|
Jul 22, 2026 |
|
Tensor Cache: Compressing Evicted Tokens into Fixed-Size Memory
|
Jul 21, 2026 |
|
MemGPT: Treating LLM Context Windows Like Virtual Memory
|
Jul 21, 2026 |
|
Lightning OPD: Precomputing the Teacher for Cheaper Reasoning Distillation
|
Jul 21, 2026 |
|
SAC: Making Sparse Attention KV Caches Actually Sparse with CXL
|
Jul 21, 2026 |
|
Coordinated MoE Scheduling: How Gimbal Fixes Expert Locality
|
Jul 21, 2026 |
|
KV Cache Eviction Through an Information Bottleneck Lens
|
Jul 21, 2026 |
|
LLMs Guess Wrong: Auditing What Models Know About You
|
Jul 21, 2026 |
|
Steering Reasoning Models' Cognitive Behaviors at Test-Time
|
Jul 21, 2026 |
|
DeltaProduct: Extending DeltaNet's State-Tracking via Householder Products
|
Jul 21, 2026 |
|
OPSDL: Teaching Long-Context Models to Trust Their Short-Context Selves
|
Jul 21, 2026 |
|
KV Cache Compaction Beyond Selection: Introducing Still
|
Jul 21, 2026 |
|
Thought Anchors: Which Sentences Really Drive LLM Reasoning
|
Jul 21, 2026 |
|
Cognitive Behaviors Behind Self-Improving Language Model Reasoners
|
Jul 20, 2026 |
|
Parcae: Stabilizing Looped Language Models with Control Theory
|
Jul 19, 2026 |
|
Cost-Aware Speculative Decoding for Mixture-of-Experts Models
|
Jul 19, 2026 |
|
Understanding Inference Scaling: Prefill, Decode, and Reasoning Bottlenecks
|
Jul 19, 2026 |
|
AMD XDNA NPU Sparsity-Aware Attention for Long-Context Prefill
|
Jul 19, 2026 |
|
HyperOffload's Scheduling Claims Under Scrutiny
|
Jul 19, 2026 |
|
SoundnessBench: Exposing AI Reviewers' Blind Spots
|
Jul 19, 2026 |
|
SkillOpt-Lite: Rethinking Agent Skill Optimization with Zeroth-Order Simplicity
|
Jul 18, 2026 |
|
Expert-Locality-Aware Decode Routing for MoE Serving
|
Jul 18, 2026 |
|
Huxley-Gödel Machine: Approximating Optimal Self-Improving Coding Agents
|
Jul 18, 2026 |
|
Do We Still Need GPUs? Rethinking AI with Matrix-Enhanced CPUs
|
Jul 18, 2026 |
|
GPT-5.6 System Card: Cybersecurity Rises, Critical Line Holds
|
Jul 18, 2026 |
|
Deep Native Structural Reasoning for Proteins, Molecules, and Crystals
|
Jul 18, 2026 |
|
Hypic: Position-Independent KV Caching for Hybrid-Attention LLM Serving
|
Jul 18, 2026 |
|
Gödel Machines: Provably Optimal Self-Rewriting AI
|
Jul 17, 2026 |
|
Darwin Gödel Machine: Self-Improving Coding Agents Through Open-Ended Evolution
|
Jul 17, 2026 |
|
SOUL.md or Severance: Host Staleness Intervention
|
Jul 14, 2026 |
|
Puzzle Compression for Hybrid MoE LLMs
|
Jul 09, 2026 |
|
Gemma 4 Open-Weight Multimodal Reasoning
|
Jul 08, 2026 |
|
When Combining Language Models Stops Helping
|
Jul 08, 2026 |
|
Why GLU Variants Improve Transformer Feed-Forward Layers
|
Jul 08, 2026 |
|
Hierarchical Sparse Attention for Infinite Context
|
Jul 08, 2026 |
|
Swish: Self-Gated Activation Beyond ReLU
|
Jul 07, 2026 |
|
Verbalizable Representations and the Global Workspace
|
Jul 07, 2026 |
|
SiLU Activations for Replay-Free Atari Reinforcement Learning
|
Jul 07, 2026 |
|
Discretizing Reward Models for RL Alignment
|
Jul 06, 2026 |
|
Program-as-Weights for Compiling Fuzzy Functions
|
Jul 06, 2026 |
|
Nemotron-TwoTower for Parallel Diffusion Language Modeling
|
Jul 05, 2026 |
|
AIConfigurator for Cross-Framework LLM Serving
|
Jul 05, 2026 |
|
SERA: Repository Specialization for Open Coding Agents
|
Jul 02, 2026 |
|
SALCA for Sparse Long-Context Decoding
|
Jul 02, 2026 |
|
DART Speeds Up Speculative LLM Decoding
|
Jul 02, 2026 |
|
When Finer Microscaling Hurts LLM Quantization
|
Jul 01, 2026 |
|
Cache-Resident LLM Inference in GB-Scale Caches
|
Jul 01, 2026 |
|
Scaling Prompt Tuning for Frozen T5 Models
|
Jul 01, 2026 |
|
Temporal-Tiered KV Cache for Long Context
|
Jun 30, 2026 |
|
SuperInfer: SLO-Aware LLM Inference on Superchips
|
Jun 30, 2026 |
|
Moebius: Seamless Parallelism Switching for MoE Serving
|
Jun 29, 2026 |
|
DSpark Improves Speculative Decoding Acceptance Rates
|
Jun 29, 2026 |
|
LLMServingSim 2.0 for Disaggregated LLM Serving
|
Jun 29, 2026 |
|
JETSPEC and Parallel Tree Speculative Decoding
|
Jun 28, 2026 |
|
Information-Aware KV Cache Compression for Long Reasoning
|
Jun 28, 2026 |
|
DAK: Direct GPU Memory Offloading for LLMs
|
Jun 27, 2026 |
|
RMSNorm: Simplifying Layer Normalization for Sequence Models
|
Jun 26, 2026 |
|
ReasonCACHE: Learning Reasoning Without Weight Updates
|
Jun 26, 2026 |
|
Prefix-Tuning for Efficient Text Generation
|
Jun 26, 2026 |
|
Learning Facts at Scale with Active Reading
|
Jun 25, 2026 |
|
RT Cores for Exact k-Nearest Neighbor Search
|
Jun 25, 2026 |
|
HELM: Holistic Evaluation of Language Models
|
Jun 25, 2026 |
|
Efficient Post-Training Quantization with FP8
|
Jun 24, 2026 |
|
AI+HW 2035: Co-Designing Efficient AI Systems
|
Jun 24, 2026 |
|
Why Open Relational Foundation Models Fail
|
Jun 24, 2026 |
|
MIOpen and AMD's Open Deep Learning Primitives
|
Jun 24, 2026 |
|
PALOMA: Benchmarking Language Model Fit Across Domains
|
Jun 24, 2026 |
|
X-LLM: Treating Multimodalities as Foreign Languages
|
Jun 23, 2026 |
|
Stable Deep RL via Gaussian Representations
|
Jun 22, 2026 |
|
Social Simulacra for Prototyping Online Communities
|
Jun 22, 2026 |
|
Fine-Tuning LLMs for Human Behavior Prediction
|
Jun 22, 2026 |
|
Building General User Models from Computer Use
|
Jun 22, 2026 |
|
Simulating Individuals with Self-Reported LLM Agents
|
Jun 22, 2026 |
|
When LeJEPA Truly Learns a World Model
|
Jun 22, 2026 |
|
TransactionGPT as a Payments Foundation Model
|
Jun 22, 2026 |
|
Modeling Financial Habits with Transaction Transformers
|
Jun 22, 2026 |
|
EMO: Emergent Modularity for Mixture-of-Experts
|
Jun 22, 2026 |
|
Weak-SIGReg for Stable Vision Transformer Training
|
Jun 20, 2026 |
|
Benchmarking PEFT Techniques for Large Language Models
|
Jun 20, 2026 |
|
Nemotron 3 Ultra for Long-Horizon Agents
|
Jun 18, 2026 |
|
SageAttention2 and Fast Exact INT4 Attention
|
Jun 18, 2026 |
|
Ling and Ring 2.6 for Trillion-Scale Agents
|
Jun 18, 2026 |
|
When Quantization Hurts Reasoning Models
|
Jun 18, 2026 |
|
InfiniGen for Efficient Long-Context LLM Inference
|
Jun 18, 2026 |
|
OpenSkill for Open-World Self-Evolution in LLM Agents
|
Jun 18, 2026 |
|
DafnyBench and LLMs for Formal Verification
|
Jun 17, 2026 |
|
seL4: Proving a Microkernel in C
|
Jun 17, 2026 |
|
DafnyPro for LLM-Assisted Dafny Verification
|
Jun 17, 2026 |
|
Training Modular KV Caches at Scale
|
Jun 17, 2026 |
|
PaperBench: Can AI Replicate AI Research?
|
Jun 17, 2026 |
|
MiniMax Sparse Attention at Million-Token Scale
|
Jun 16, 2026 |
|
From AGI to ASI and Beyond
|
Jun 16, 2026 |
|
ACE: Matrix-Native AI Extensions for x86
|
Jun 15, 2026 |
|
Dafny for Trustworthy AI Code Generation
|
Jun 15, 2026 |
|
When LoRA Helps Under KV Cache Compression
|
Jun 14, 2026 |
|
KV Binding Is Secretly Linear Attention
|
Jun 14, 2026 |
|
Can LLMs Enable Mainstream Formal Verification?
|
Jun 14, 2026 |
|
From Natural Language to Verified Dafny Code
|
Jun 14, 2026 |
|
AllMem for Efficient Long-Context Modeling
|
Jun 13, 2026 |
|
IndexMem: Learned KV-Cache Eviction for Long-Context LLMs
|
Jun 13, 2026 |
|
Lattice: Fixed-Slot Compression for Transformer Memory
|
Jun 12, 2026 |
|
Relational Graph Transformer for Multi-Table Learning
|
Jun 12, 2026 |
|
Learning at Test Time with Expressive RNN States
|
Jun 11, 2026 |
|
Atlas: Test-Time Memory for Long Contexts
|
Jun 11, 2026 |
|
KumoRFM for In-Context Relational Learning
|
Jun 11, 2026 |
|
Robots Need More Than VLAs and World Models
|
Jun 11, 2026 |
|
Do Transformers Need Three Projections?
|
Jun 11, 2026 |
|
KumoRFM-2 for Relational Learning at Scale
|
Jun 10, 2026 |
|
Unembedding Matrices as Feature Lenses for Embeddings
|
Jun 10, 2026 |
|
End-to-End Context Compression at Scale
|
Jun 10, 2026 |
|
Latent Reasoning with Normalizing Flows
|
Jun 10, 2026 |
|
RAGEN-2: Reasoning Collapse in Agentic RL
|
Jun 10, 2026 |
|
Unified Neural Scaling Laws Across Regimes
|
Jun 10, 2026 |
|
Predictive Query Language for Relational Databases
|
Jun 10, 2026 |
|
EMO: Emergent Modularity in Sparse Language Models
|
Jun 08, 2026 |
|
Harvest: Borrowing Peer GPU Memory for LLMs
|
Jun 05, 2026 |
|
VeriCache: Lossless LLM Inference from Lossy KV Caches
|
Jun 05, 2026 |
|
SFMP Search-Free Mixed-Precision LLM Quantization
|
Jun 05, 2026 |
|
Technical AGI Safety and Security Framework
|
Jun 05, 2026 |
|
Mooncake for KV Cache-Centric LLM Serving
|
Jun 05, 2026 |
|
When AI Builds Itself and Recursive Self-Improvement
|
Jun 05, 2026 |
|
Opaque Serial Depth and Chain-of-Thought Limits
|
Jun 03, 2026 |
|
Memory-Bound, Not Bandwidth-Limited Batch-1 LLM Decode
|
Jun 02, 2026 |
|
TRELLIS and Bounded-Memory Transformer KV Compression
|
Jun 02, 2026 |
|
Post-Trained MoE Skips Half Its Experts
|
Jun 01, 2026 |
|
SmolLM2 and the Power of Better Data
|
Jun 01, 2026 |
|
Dragonfly Topology for Scalable AI Networks
|
Jun 01, 2026 |
|
Do Language Models Need Sleep?
|
Jun 01, 2026 |
|
Snap's Microkernel Approach to Host Networking
|
Jun 01, 2026 |
|
KVzap: Fast, Adaptive, Faithful KV Cache Pruning
|
May 30, 2026 |
|
KVzip for Query-Agnostic KV Cache Compression
|
May 29, 2026 |
|
DFX: Multi-FPGA Acceleration for Transformer Inference
|
May 27, 2026 |
|
Trajectory Summaries for Long-Horizon Coding Agents
|
May 27, 2026 |
|
Beluga: CXL Memory Pooling for LLM KV Cache
|
May 27, 2026 |
|
CXL-GPU and Beyond Onboard Memory
|
May 27, 2026 |
|
Generative Recursive Reasoning in Latent Space
|
May 26, 2026 |
|
LPU Chip for Low-Latency LLM Inference
|
May 23, 2026 |
|
After Titans: Behrouz on Nested Learning and Hope
|
May 22, 2026 |
|
Titans: Learning to Memorize at Test Time
|
May 20, 2026 |
|
The Sparsity Wall: What Reiner Pope Told Dwarkesh About MoE and Sparse Attention
|
May 19, 2026 |
|
Serving MoE Models with Disaggregated Expert Parallelism
|
May 19, 2026 |
|
Affordable Large-Scale Decoding Through Model-System Co-Design
|
May 19, 2026 |
|
NanoFlow and the Future of LLM Serving
|
May 19, 2026 |
|
Air Force One, Jensen Huang, and Anthropic's 2028 Memo
|
May 15, 2026 |
|
Deep Kernel Fusion for Transformer Decoding
|
May 15, 2026 |
|
FlashFuser and Hopper-Era FFN Kernel Fusion
|
May 15, 2026 |
|
Lossless Sparse Deltas for RL Networks
|
May 15, 2026 |
|
JANUS for Scalable MoE Inference
|
May 15, 2026 |
|
Scaling Laws for Multilingual Code Pretraining
|
May 15, 2026 |
|
Causal-JEPA for Object-Level World Models
|
May 15, 2026 |
|
Ministral 3: Cascade Distillation for Long-Context Multimodal Models
|
May 15, 2026 |
|
When Many-Shot CoT Becomes Test-Time Learning
|
May 15, 2026 |
|
Agentic AI as a Path to AGI
|
May 15, 2026 |
|
Trace Rewriting Against Unauthorized LLM Distillation
|
May 15, 2026 |
|
TMAS: Scaling Test-Time Compute with Multi-Agent Synergy
|
May 14, 2026 |
|
AI Co-Mathematician for Mathematical Research
|
May 14, 2026 |
|
δ-mem and Online Memory for LLMs
|
May 13, 2026 |
|
Qwen-Image-2.0 for Unified Generation and Editing
|
May 13, 2026 |
|
MiA-Signature and Global Activation for Long Context
|
May 13, 2026 |
|
Long Context Pre-Training with Lighthouse Attention
|
May 13, 2026 |
|
MELT: Decoupling Compute From Memory
|
May 13, 2026 |
|
ELF and Continuous Language Diffusion
|
May 12, 2026 |
|
ForkKV for Multi-LoRA Agent Serving
|
May 12, 2026 |
|
Metacognition Against Confident Hallucinations
|
May 12, 2026 |
|
TIDE and the Rare Token Problem
|
May 12, 2026 |
|
Agentic Discovery for Test-Time Scaling
|
May 12, 2026 |
|
Vistara Brings CXL Memory to Hyperscale
|
May 11, 2026 |
|
Split Personality Training Reveals Latent Knowledge
|
May 10, 2026 |
|
Why Transformers Fail at Counting
|
May 10, 2026 |
|
Optimization, Credit Assignment, and Consciousness
|
May 10, 2026 |
|
Reasoning Theater and Unfaithful Chain-of-Thought
|
May 10, 2026 |
|
RAPTOR: Stable Concept Directions From Logistic Probes
|
May 10, 2026 |
|
When LLM Judges Become Coin Flips
|
May 09, 2026 |
|
LAPS for Length-Aware LLM Serving
|
May 09, 2026 |
|
Generative Modeling via Drifting in One Step
|
May 09, 2026 |
|
EverMemOS for Long-Horizon Agent Memory
|
May 09, 2026 |
|
Explicit Information Transmission for Context Compression
|
May 09, 2026 |
|
Why LLM Serving Needs Mathematical Optimization
|
May 09, 2026 |
|
Can Models Learn from Long Context?
|
May 08, 2026 |
|
How Models Detect Hidden Activation Steering
|
May 08, 2026 |
|
What the Frog’s Eye Tells the Brain
|
May 08, 2026 |
|
Learning in Random Nets and Generalization
|
May 08, 2026 |
|
Backpropagation Through Time Explained
|
May 08, 2026 |
|
SGLang for Faster Structured LLM Programs
|
May 08, 2026 |
|
TensorFlow for Distributed Machine Learning Systems
|
May 08, 2026 |
|
Machine Learning Self-Calibrated FPGA Time-to-Digital Converter
|
May 08, 2026 |
|
Why LightGBM Made Boosted Trees Fast
|
May 07, 2026 |
|
Fast FPGA BDT Inference for LHC Triggers
|
May 07, 2026 |
|
Fast FPGA Inference for LHC Triggers
|
May 07, 2026 |
|
Boosted Decision Trees for CMS Muon Triggers
|
May 07, 2026 |
|
Automating DNN Compilation for FPGA Accelerators
|
May 07, 2026 |
|
Synchronous Data Flow for Signal Processing
|
May 07, 2026 |
|
Caffeine: A Unified FPGA for CNNs
|
May 07, 2026 |
|
Caffe and the Rise of CNN Frameworks
|
May 07, 2026 |
|
RFNoC SISO Processor via High-Level Synthesis
|
May 06, 2026 |
|
Automating CNN Mapping on Embedded FPGAs
|
May 06, 2026 |
|
Training Million-Token LLMs Beyond the Memory Barrier
|
May 04, 2026 |
|
Training LLMs for Divide-and-Conquer Reasoning
|
May 04, 2026 |
|
Reinforcement Learning in 2025: An Overview
|
May 04, 2026 |
|
PackKV Lossy Compression for KV Caches
|
May 04, 2026 |
|
Geometric Memory in Deep Sequence Models
|
May 03, 2026 |
|
Selective Classification with Deep Neural Networks
|
May 03, 2026 |
|
Self-Improving Pretraining With Post-Trained Models
|
May 03, 2026 |
|
How Induction Heads Emerge in Transformers
|
May 03, 2026 |
|
node2vec and Learning Graph Embeddings
|
May 03, 2026 |
|
DeepWalk and the Rise of Graph Embeddings
|
May 03, 2026 |
|
Fast Speech Recognition by Transcript Editing
|
May 02, 2026 |
|
Discrete Representations for Continual Reinforcement Learning
|
May 02, 2026 |
|
Recursive Multi-Agent Systems in Latent Space
|
May 02, 2026 |
|
A Practical Review of Mechanistic Interpretability
|
May 02, 2026 |
|
DeltaKV: Compressing KV Caches for Long Context
|
May 02, 2026 |
|
CacheFlow and 3D-Parallel KV Cache Restoration
|
May 02, 2026 |
|
Do Language Models Know Their Limits
|
May 02, 2026 |
|
Hopfield Networks and Transformer Attention as Memory
|
May 01, 2026 |
|
Sequence to Sequence Learning with Neural Networks
|
May 01, 2026 |
|
Bahdanau Attention for Neural Machine Translation
|
May 01, 2026 |
|
Convolutional Sequence to Sequence Learning for Translation
|
May 01, 2026 |
|
Teaching Language Models to Verbalize Uncertainty
|
May 01, 2026 |
|
Can LLMs Judge Their Own Capabilities?
|
May 01, 2026 |
|
ChartNet for Robust Multimodal Chart Understanding
|
May 01, 2026 |
|
Scaling Test-Time Compute for Reasoning Models
|
Apr 30, 2026 |
|
Breaking the Prefix Barrier with Shared KV Cache
|
Apr 30, 2026 |
|
Reverse-Mode Differentiation Across AD and Neural Nets
|
Apr 30, 2026 |
|
Stabilizing Efficient Reasoning with Step-Level Advantage Selection
|
Apr 29, 2026 |
|
Stochastic KV Routing for Cache Sharing
|
Apr 29, 2026 |
|
Deep Learning in Spiking Neural Networks
|
Apr 28, 2026 |
|
FPGA Neural Network Accelerators for Space
|
Apr 28, 2026 |
|
World-R1 Improves 3D Consistency in Text-to-Video
|
Apr 28, 2026 |
|
Long Short-Term Memory and Vanishing Gradients
|
Apr 27, 2026 |
|
Compressed Convolutional Attention in Latent Space
|
Apr 27, 2026 |
|
AgenticQwen and Small Industrial Tool Agents
|
Apr 27, 2026 |
|
Experience-Based Learning Beyond Human Data
|
Apr 26, 2026 |
|
Breiman's Two Cultures of Statistical Modeling
|
Apr 26, 2026 |
|
Muon Is Scalable for LLM Training
|
Apr 25, 2026 |
|
Kimi K2.5 and Visual Agent Swarms
|
Apr 25, 2026 |
|
ScoutAttention for Efficient KV Cache Offloading
|
Apr 25, 2026 |
|
DeepSeek-V4 and Practical Million-Token Context
|
Apr 25, 2026 |
|
Agentic Aggregation for Long-Horizon AI Tasks
|
Apr 23, 2026 |
|
TokenDance for Multi-Agent KV Cache Sharing
|
Apr 23, 2026 |
|
RetrievalAttention for Long-Context LLM Inference
|
Apr 22, 2026 |
|
DreamerV3 World Models Across 150 Tasks
|
Apr 22, 2026 |
|
Program Synthesis with Large Language Models
|
Apr 22, 2026 |
|
Test-time Scaling for Multi-Agent Collaborative Reasoning
|
Apr 22, 2026 |
|
TUMIX Multi-Agent Test-Time Scaling with Tools
|
Apr 22, 2026 |
|
Benchmarking Test-Time Scaling for General LLM Agents
|
Apr 22, 2026 |
|
Distilling Multi-Agent Reasoning into a Single LLM
|
Apr 22, 2026 |
|
Efficient KV Cache Sharing for Multi-LoRA Agents
|
Apr 22, 2026 |
|
CacheBlend for Fast RAG Serving
|
Apr 22, 2026 |
|
Directly Trained Spiking DQNs for Atari
|
Apr 22, 2026 |
|
Qwen3.5-Omni Thinker-Talker for Omnimodal Streaming
|
Apr 21, 2026 |
|
The Abstraction Fallacy and AI Consciousness
|
Apr 20, 2026 |
|
ContiguousKV for Faster LLM Prefill KV Reuse
|
Apr 20, 2026 |
|
Gated Delta Networks for Long-Context Retrieval
|
Apr 20, 2026 |
|
Prefill-as-a-Service for Cross-Datacenter KV Cache
|
Apr 20, 2026 |
|
Nemotron 3 Super Hybrid Mamba-Transformer MoE
|
Apr 20, 2026 |
|
Parallelizing DeltaNet Linear Transformers over Sequence Length
|
Apr 19, 2026 |
|
Gated Linear Attention for Efficient Long Sequences
|
Apr 19, 2026 |
|
Defining AGI as Adaptive Artificial Scientist
|
Apr 19, 2026 |
|
Mamba-3 for Efficient Sequence Modeling
|
Apr 16, 2026 |
|
Linear Classifier Probes for Intermediate Layers
|
Apr 16, 2026 |
|
KVSwap for Disk-Aware Long-Context On-Device Inference
|
Apr 16, 2026 |
|
Neural Chameleons and Evading Activation Monitors
|
Apr 15, 2026 |
|
SkillsBench for Evaluating Agent Skills
|
Apr 15, 2026 |
|
Experimental Comparison of Agentic and Enhanced RAG
|
Apr 14, 2026 |
|
Learning to Reason with 13 Parameters
|
Apr 14, 2026 |
|
Memory Intelligence Agents for Deep Research
|
Apr 13, 2026 |
|
GPU-Accelerated Dynamic Quantized ANNS Graph Search
|
Apr 13, 2026 |
|
Latent Space as a New Computational Paradigm
|
Apr 12, 2026 |
|
Learning Latent Action World Models from Video
|
Apr 12, 2026 |
|
KV Cache TTL for Multi-Turn Agent Scheduling
|
Apr 12, 2026 |
|
VL-JEPA for Vision-Language Semantic Prediction
|
Apr 12, 2026 |
|
ClawBench for Real-World Online AI Agents
|
Apr 12, 2026 |
|
FengHuang for Rack-Scale LLM Inference Memory
|
Apr 12, 2026 |
|
ASI-Evolve for Data, Architectures, and RL
|
Apr 12, 2026 |
|
In-Place Test-Time Training for Transformers
|
Apr 11, 2026 |
|
Neural Computers as Learned Latent Runtimes
|
Apr 11, 2026 |
|
DRAM-Free In-Flash Computing for LLM Inference
|
Apr 10, 2026 |
|
Computation-Bandwidth-Memory Trade-offs for AI Infrastructure
|
Apr 10, 2026 |
|
TriAttention for Efficient Long-Context KV Compression
|
Apr 08, 2026 |
|
When Spectral Gradient Updates Help Deep Learning
|
Apr 08, 2026 |
|
Memory Sparse Attention for 100M-Token Scaling
|
Apr 08, 2026 |
|
Cache Mechanism for Agent RAG Systems
|
Apr 08, 2026 |
|
Speculative Decoding in Real vLLM Serving
|
Apr 07, 2026 |
|
Real Context Size and Context Rot
|
Apr 07, 2026 |
|
MEMSEARCHER: Reinforcement Learning for LLM Memory Management
|
Apr 04, 2026 |
|
Recursive Language Models for Arbitrarily Long Prompts
|
Apr 04, 2026 |
|
Internal Safety Collapse in Frontier LLMs
|
Apr 04, 2026 |
|
Kosmos AI Scientist for Autonomous Discovery
|
Apr 04, 2026 |
|
QVCache for Semantic Caching in ANN Search
|
Apr 04, 2026 |
|
FlatAttention for Tile-Based Accelerator Inference
|
Apr 04, 2026 |
|
IMO-Bench for Robust Mathematical Reasoning
|
Apr 04, 2026 |
|
Batch-Aware Expert Routing for Faster MoE Decoding
|
Apr 03, 2026 |
|
CXL Computational Memory Offloading for Lower Runtime
|
Apr 03, 2026 |
|
OOD Shifts Make LLM Representations Sparser
|
Apr 02, 2026 |
|
Meta-Harness and the Power of LLM Plumbing
|
Apr 02, 2026 |
|
Emergent Social Risks in Multi-Agent Systems
|
Apr 02, 2026 |
|
AI Agent Traps and Prompt Injection
|
Apr 02, 2026 |
|
Simple Self-Distillation for Better Code Generation
|
Apr 02, 2026 |
|
MetaClaw: Just Talk and Continual Agent Adaptation
|
Mar 31, 2026 |
|
MAML and the Basics of Meta-Learning
|
Mar 30, 2026 |
|
Doc-to-LoRA: Internalizing Context as LoRA
|
Mar 30, 2026 |
|
Agentic AI and the Next Intelligence Explosion
|
Mar 29, 2026 |
|
TurboQuant: Online Vector Quantization with Near-optimal Distortion Rate
|
Mar 28, 2026 |
|
Splitwise: Phase-Split LLM Inference
|
Mar 28, 2026 |
|
From Prefix Cache to Fusion RAG Cache: Accelerating LLM Inference in Retrieval-Augmented Generation
|
Mar 27, 2026 |
|
HyperAgents and Metacognitive Self-Improvement
|
Mar 27, 2026 |
|
LeWorldModel: Stable Joint-Embedding World Models from Pixels
|
Mar 26, 2026 |
|
LookaheadKV: Fast and Accurate KV Cache Eviction by Glimpsing into the Future without Generation
|
Mar 26, 2026 |
|
Language Models are Injective and Hence Invertible
|
Mar 25, 2026 |
|
Jet-Nemotron and PostNAS for Faster Long Context
|
Mar 25, 2026 |
|
Speculative Speculative Decoding
|
Mar 25, 2026 |
|
Lookahead Q-Cache for Consistent KV Eviction
|
Mar 25, 2026 |
|
New Site Features: Conferences Tracking and How to Support the Show
|
Mar 20, 2026 |
|
Accelerating LLM Cold Starts with Programmable Page Cache
|
Mar 19, 2026 |
|
SolidAttention: Co-Designing Sparse Attention and SSD I/O
|
Mar 19, 2026 |
|
Generative File Systems: Replacing Code with Formal Specifications
|
Mar 19, 2026 |
|
Xerxes: CXL 3.0 Simulation for Scalable Memory Systems
|
Mar 19, 2026 |
|
Optimizing Mixture of Block Attention Through Statistical Theory
|
Mar 19, 2026 |
|
Predicting Task Performance with Context-aware Scaling Laws
|
Mar 17, 2026 |
|
Model-Aware Tokenizer Transfer for Multilingual LLMs
|
Mar 17, 2026 |
|
xLLM: Co-Locating Online and Offline LLM Inference
|
Mar 17, 2026 |
|
Memory in the Age of AI Agents: Forms, Functions, Dynamics
|
Mar 17, 2026 |
|
Qwen3Guard: Streaming Three-Way Safety Classification for LLMs
|
Mar 17, 2026 |
|
Bidaw: Computation-Storage Aware KV Caching for LLMs
|
Mar 17, 2026 |
|
New Voices, Same Nerds: The Kokoro TTS Episode
|
Mar 17, 2026 |
|
Rethinking Residual Connections: Learned Attention Across Transformer Depth
|
Mar 17, 2026 |
|
CacheSlide: Position-Aware KV Cache Reuse for Agent LLMs
|
Mar 17, 2026 |
|
The Art of Scaling Reinforcement Learning Compute for LLMs
|
Mar 16, 2026 |
|
LLM Agents Reason About Code Without Running It
|
Mar 15, 2026 |
|
Emergent Cooperation in Self-Interested Multi-Agent AI
|
Mar 13, 2026 |
|
50x KV Cache Compression in Seconds via Attention Matching
|
Mar 12, 2026 |
|
Gradient Descent at Inference Time for LLM Reasoning
|
Mar 11, 2026 |
|
MalGEN: Multi-Agent AI for Red Teaming Malware
|
Mar 08, 2026 |
|
Structured State Space Duality Unifies Transformers and SSMs
|
Mar 07, 2026 |
|
DualPath Breaks Storage Bandwidth Bottleneck in Agentic Inference
|
Mar 07, 2026 |
|
Why CARTRIDGE Works: Keys as Routers in KV Caches
|
Mar 07, 2026 |
|
Systematic Characterization of LLM Inference on GPUs
|
Mar 07, 2026 |
|
Tokenization Bias: The Hidden Flaw Breaking Language Models
|
Mar 07, 2026 |
|
NVIDIA Nemotron 3 Hybrid SSM Transformer Architecture
|
Mar 07, 2026 |
|
We're Open Source! New Home, Visualizations, and How to Shape Our Queue
|
Mar 06, 2026 |
|
FlashAttention-4 Conquers Asymmetric GPU Hardware Scaling
|
Mar 06, 2026 |
|
Parallel Token Prediction: From ProphetNet to Dependent Multi-Token Generation
|
Mar 04, 2026 |
|
Unified Latents (UL): How to train your latents
|
Feb 28, 2026 |
|
QuantSpec: Hierarchical KV Cache for Self-Speculative Decoding
|
Feb 28, 2026 |
|
MatFormer: Nested Transformer for Elastic Inference
|
Feb 28, 2026 |
|
MagicDec: Breaking Latency-Throughput Tradeoffs via KV-Compressed Speculative Decoding
|
Feb 28, 2026 |
|
KV selection algorithms: static (SnapKV) Vs dynamic (PQCache)
|
Feb 28, 2026 |
|
Fast Inference from Transformers via Speculative Decoding
|
Feb 28, 2026 |
|
EAGLE: Evolution of Lossless Acceleration for LLM Inference
|
Feb 28, 2026 |
|
CXL-SpecKV: Bridging the LLM Memory Wall with Speculative FPGA Disaggregation
|
Feb 28, 2026 |
|
Building Production-Ready Speculative Decoding with TensorRT-LLM
|
Feb 28, 2026 |
|
Apple's Speculative Streaming: Fast LLM Inference without Auxiliary Models
|
Feb 28, 2026 |
|
Apple's Mirror Speculative Decoding: Parallel LLM Inference via Heterogeneous Accelerators
|
Feb 28, 2026 |
|
Adaptive Control for Batched Speculative Decoding in LLM Serving
|
Feb 28, 2026 |
|
Taming the Long-Tail: Efficient Reasoning RL with Adaptive Drafters
|
Feb 26, 2026 |
|
Optimizing Verification and Efficiency in Multi-Draft Speculative Decoding
|
Feb 26, 2026 |
|
MEDUSA: Parallel Decoding Heads for Accelerated LLM Inference
|
Feb 26, 2026 |
|
Measuring LLM Reasoning Effort via Deep-Thinking Tokens
|
Feb 26, 2026 |
|
Measuring AI Ability to Complete Long Tasks
|
Feb 26, 2026 |
|
FastGRPO: Concurrency-Aware Speculative Decoding for Policy Optimization
|
Feb 26, 2026 |
|
Evaluating Collective Behaviour of Hundreds of LLM Agents
|
Feb 26, 2026 |
|
Draft & Verify: Lossless Large Language Model Acceleration via Self-Speculative Decoding
|
Feb 26, 2026 |
|
Deep Learning Frameworks for Robust Quadrupedal Locomotion
|
Feb 26, 2026 |
|
Advancements in Efficient KV Cache Quantization and Management
|
Feb 26, 2026 |
|
Accelerating Large Language Model Decoding
with Speculative Sampling
|
Feb 26, 2026 |
|
NeurIPS 2025: Agentic Plan Caching: Test-Time Memory for Fast and Cost-Efficient LLM Agents
|
Feb 25, 2026 |
|
FAST26: Bidaw: Enhancing Key-Value Caching for Interactive LLM Serving via Bidirectional
|
Feb 25, 2026 |
|
FAST26: CacheSlide: Unlocking Cross Position-Aware KV Cache Reuse for Accelerating LLM Serving
|
Feb 25, 2026 |
|
Cortex: Semantic Knowledge Caching for Low-Latency LLM Agents
|
Feb 25, 2026 |
|
Agentic Context Engineering: Evolving Contexts for Self-Improving Language Models
|
Feb 25, 2026 |
|
1.3 Billion Agents by 2028: The $50 Billion Boom and the Hidden Enterprise Crisis
|
Feb 25, 2026 |
|
Zero-Shot Context Generalization in Reinforcement
Learning from Few Training Contexts
|
Feb 24, 2026 |
|
Superintelligent Agents Pose Catastrophic Risks:
Can Scientist AI Offer a Safer Path?
|
Feb 24, 2026 |
|
PPDS: Achieving Persona Consistency Through Large-Scale Dialogue Data Engineering
|
Feb 24, 2026 |
|
Probing Scientific General Intelligence of LLMs
with Scientist-Aligned Workflows
|
Feb 24, 2026 |
|
Personalized Dialogue Generation via Persona-Adaptive Attention
|
Feb 24, 2026 |
|
PersonaPKT: Parameter-Efficient Knowledge Transfer for Personalized Dialogue Agents
|
Feb 24, 2026 |
|
Persona Vectors: monitoring and controlling character traits on LLMs
|
Feb 24, 2026 |
|
Machine Learning for Electrophysiological Phenotyping of Schizophrenia and Bipolar Disorder
|
Feb 24, 2026 |
|
Evaluating LLM Embeddings for Psychometric Personality Prediction
|
Feb 24, 2026 |
|
Bloom: an open source tool for automated behavioral evaluations
|
Feb 24, 2026 |
|
AI and the Decline of Entry-Level Employment
|
Feb 24, 2026 |
|
2019 UNILM: Unified Language Model Pre-training for NLU and NLG
|
Feb 24, 2026 |
|
Procgen Benchmark: Measuring Generalization in Reinforcement Learning
|
Feb 20, 2026 |
|
Learning beyond Teacher: Generalized On-Policy Distillation with Reward Extrapolation
|
Feb 20, 2026 |
|
GLM-5: Transitioning from Vibe Coding to Agentic Engineering
|
Feb 20, 2026 |
|
Experiential Reinforcement Learning: Internalizing Reflection for Better Policy Training
|
Feb 20, 2026 |
|
Contrastive Behavioral Similarity Embeddings for Generalization in Reinforcement Learning
|
Feb 20, 2026 |
|
A 2024 Survey Analyzing Generalization in
Deep Reinforcement Learning
|
Feb 20, 2026 |
|
Voxtral Realtime: Native Streaming ASR with Sub-Second Latency
|
Feb 17, 2026 |
|
The Endless Gym: Training Terminal Agents
|
Feb 17, 2026 |
|
Teaching Models to Teach Themselves via Stepping Stone Curricula
|
Feb 17, 2026 |
|
Moltbook: The Heartbeat of Autonomy: Fingerprinting Human Influence in AI Societies
|
Feb 17, 2026 |
|
Jet-RL: Stable On-Policy Reinforcement Learning with Unified FP8 Flow
|
Feb 17, 2026 |
|
Information Bottleneck-based Causal Attention for Medical Image Recognition
|
Feb 17, 2026 |
|
Intelligent AI Delegation
|
Feb 17, 2026 |
|
DeepVerifier: Self-Evolving Research Agents via Rubric-Guided Verification
|
Feb 17, 2026 |
|
Advancing Mechanistic Interpretability with Sparse Autoencoders
|
Feb 17, 2026 |
|
Adaptive Test-Time Scaling with World Models for Visual Spatial Reasoning
|
Feb 17, 2026 |
|
Outcome Accuracy is Not Enough: Aligning the Reasoning Process of Reward Models
|
Feb 13, 2026 |
|
Sapient Intelligence: Hierarchical Reasoning Model
|
Feb 11, 2026 |
|
Reinforced Attention Learning
|
Feb 11, 2026 |
|
LongCat: Scaling Embeddings Outperforms Scaling Experts in Language Models
|
Feb 11, 2026 |
|
DR. KERNEL: Reinforcement Learning for Optimized Triton Kernel Generation
|
Feb 11, 2026 |
|
Dario Amodei: The Adolescence of Technology
|
Feb 11, 2026 |
|
Dario Amodei: Machines of Loving Grace
|
Feb 11, 2026 |
|
ChunkKV: Semantic-Preserving KV Cache
Compression for Efficient Long-Context LLM
Inference
|
Feb 11, 2026 |
|
Advances in Attention Distillation for Efficient Transformer Models
|
Feb 11, 2026 |
|
A Comprehensive Survey of Mixture-of-Experts:
Algorithms, Theory, and Applications
|
Feb 11, 2026 |
|
Towards a Science of Scaling Agent Systems
|
Feb 09, 2026 |
|
Uncertainty-aware genomic deep learning with knowledge distillation
|
Feb 06, 2026 |
|
Reinforcement Learning via Self-Distillation
|
Feb 06, 2026 |
|
On-Policy Self-Distillation for Advanced LLM Reasoning
|
Feb 06, 2026 |
|
Moloch’s Bargain: Market Incentives and the Rise of AI Misalignment
|
Feb 06, 2026 |
|
Knowledge distillation to context distillation
|
Feb 06, 2026 |
|
Distilling GNN Knowledge into Non-Neural Cell Graph Student Models
|
Feb 06, 2026 |
|
DeepSearchQA: Bridging the Comprehensiveness Gap for Deep Research Agents
|
Feb 06, 2026 |
|
Claude Opus 4.6 Technical Report and Agent Capabilities
|
Feb 06, 2026 |
|
Advancing regulatory variant effect prediction with AlphaGenome
|
Feb 06, 2026 |
|
2006 Model Compression: Ensembles
|
Feb 06, 2026 |
|
2015: Distilling the Knowledge in a Neural Network
|
Feb 06, 2026 |
|
Reasoning Models Generate Societies of Thought
|
Jan 28, 2026 |
|
NeurIPS 2025: L2M: Mutual Information Scaling Law for Long-Context Language Modeling
|
Jan 28, 2026 |
|
Long context: Dichotomy of Findings & Status of Research
|
Jan 28, 2026 |
|
Keel: Post-LayerNorm Is Back: Stable, ExpressivE, and Deep
|
Jan 28, 2026 |
|
SLDAgent: Evolutionary Discovery of Superhuman AI Scaling Laws
|
Jan 26, 2026 |
|
Sequoia Capital: AGI is here
|
Jan 24, 2026 |
|
OpenAI: Scaling PostgreSQL to 800 Million ChatGPT Users
|
Jan 24, 2026 |
|
Agentic Reasoning for Large Language Models: A Comprehensive Roadmap
|
Jan 24, 2026 |
|
MEMRL: Self-Evolving Agents via Runtime Reinforcement Learning on Episodic
|
Jan 23, 2026 |
|
Google: R&D inference value on HBF + PNM + low latency interconnect
|
Jan 23, 2026 |
|
Storage-next: Do We Need New Hardware for AI Storage, or Just Better Layouts?
|
Jan 21, 2026 |
|
Meta's solution to massive DLRM inference through software defined memory
|
Jan 21, 2026 |
|
LeCun's AMI Energy-Based Models and the Path to Autonomous Intelligence
|
Jan 21, 2026 |
|
Process Reward Learning for LLM Reasoning Optimization
|
Jan 19, 2026 |
|
Parallel Context-of-Experts Decoding for Efficient RAG reasoning
|
Jan 19, 2026 |
|
OpenRouter 2025 Report: Analysis of Global LLM Usage Patterns
|
Jan 19, 2026 |
|
Multidimensional Safety Evaluation of Frontier AI Models
|
Jan 19, 2026 |
|
MemoBrain: Executive Memory for Tool-Augmented Reasoning Agents
|
Jan 19, 2026 |
|
MATTRL: Collaborative Test-Time Reinforcement Learning for Multi-Agent Reasoning
|
Jan 19, 2026 |
|
How to make LLMs better at sci-fi writing
|
Jan 19, 2026 |
|
H-net: End-to-End Hierarchical Sequence Modeling via Dynamic Chunking
|
Jan 19, 2026 |
|
GLM-Image: Towards
Versatile Multimodal Reasoning with Scalable
Reinforcement Learning
|
Jan 19, 2026 |
|
CompassMem: Event-Centric Logic Maps for Agent Memory
|
Jan 19, 2026 |
|
A Chain of Thought reasoning academic brawl
|
Jan 19, 2026 |
|
Squisher: Approximating the Fisher Information Matrix and use cases
|
Jan 17, 2026 |
|
Scaling laws: long context length and in context learning
|
Jan 17, 2026 |
|
NVIDIA: TTT-E2E: Unlocking Long-Context Learning via End-to-End Test-Time Training
|
Jan 17, 2026 |
|
Attention with a bias
|
Jan 17, 2026 |
|
DeepSeek Engram: Scaling Large Language Models via Conditional Memory Lookup
|
Jan 14, 2026 |
|
PageANN: Scalable Disk ANNS with Page-Aligned Graphs
|
Dec 07, 2025 |
|
NeurIPS 2025: Homogeneous Keys, Heterogeneous Values
|
Dec 04, 2025 |
|
NeurIPS 2025: MoBA: Mixture of Block Attention for Long-Context LLMs
|
Nov 29, 2025 |
|
NeurIPS 2025: Reward Reasoning Model
|
Nov 29, 2025 |
|
NeurIPS 2025: Parallel Scaling Law for Language Models
|
Nov 29, 2025 |
|
NeurIPS 2025: Thinkless: LLM Learns When to Think
|
Nov 29, 2025 |
|
NeurIPS 2025: SeRL: Self-Play Reinforcement Learning for Large Language Models with Limited Data
|
Nov 29, 2025 |
|
NeurIPS 2025: Self-Adapting Language Models
|
Nov 29, 2025 |
|
NeurIPS 2025: Reinforcement Learning for Reasoning in Large Language Models with One Training Example
|
Nov 29, 2025 |
|
NeurIPS 2025: KGGen: Extracting Knowledge Graphs from Plain Text with Language Models
|
Nov 29, 2025 |
|
NeurIPS 2025: Large Language Diffusion Models
|
Nov 29, 2025 |
|
NeurIPS 2025: DYNAACT: Large Language Model Reasoning with Dynamic Action Spaces
|
Nov 29, 2025 |
|
NeurIPS 2025: Gated Attention for Large Language Models: Non-linearity, Sparsity, and Attention-Sink-Free
|
Nov 29, 2025 |
|
NeurIPS 2025: FlashBias: Fast Computation of Attention with Bias
|
Nov 29, 2025 |
|
NeurIPS 2025: A-Mem: Agentic Memory for LLM Agents
|
Nov 29, 2025 |
|
Anthropic: Disrupting the First AI-Orchestrated Cyber Espionage Campaign
|
Nov 27, 2025 |
|
Neuromorphic computing: Brain-Inspired AI and Hardware
|
Nov 22, 2025 |
|
DeepSeek-OCR: Contexts Optical Compression
|
Nov 22, 2025 |
|
Anthropic: reward hacking & misalignment & sabotage
|
Nov 22, 2025 |
|
Meta: SAM 3
|
Nov 20, 2025 |
|
vAttention Vs Strata: advanced GPU memory management
|
Nov 19, 2025 |
|
MLP Mixer Models
|
Nov 19, 2025 |
|
Mixture-of-Depths: Dynamic Compute Allocation in Transformers
|
Nov 19, 2025 |
|
Mamba-360: State Space Models for Long Sequence Modeling
|
Nov 19, 2025 |
|
Marin: Open LLM Optimization & Diagnostics
|
Nov 19, 2025 |
|
AMD: Instella: Fully Open Language Models with Stellar Performance
|
Nov 16, 2025 |
|
Mechanistic interpretability: Decoding the AI's Inner Logic: Circuits and Sparse Features
|
Nov 15, 2025 |
|
Vidur: Simulation for Efficient LLM Inference Deployment
|
Nov 10, 2025 |
|
SYMPHONY: Memory Management for LLM Multi-Turn Inference
|
Nov 10, 2025 |
|
Tempo: SLO-Aware LLM Serving Maximizing Service Gain
|
Nov 10, 2025 |
|
Spectral Gap: Analysis of Attention Layers and Graph Transformers
|
Nov 10, 2025 |
|
Random Walk Methods for Graph Learning and Networks
|
Nov 10, 2025 |
|
Metacognition and Skill Discovery in LLM Math Reasoning
|
Nov 10, 2025 |
|
LLM-AutoDiff: Auto-Differentiate Any LLM Workflow
|
Nov 10, 2025 |
|
DSPy and TextGrad: Compiling Language Model Systems
|
Nov 10, 2025 |
|
Doubly Stochastic Attention for Transformers
|
Nov 10, 2025 |
|
Continuous Autoregressive Language Models: CALM
|
Nov 10, 2025 |
|
Context Distillation for Language Models
|
Nov 10, 2025 |
|
Confucius: Intent-Driven Network Management with Multi-Agent LLMs
|
Nov 10, 2025 |
|
CARTRIDGE: Efficient In-Context Learning via Distillation
|
Nov 10, 2025 |
|
AlphaEvolve: Mathematical Discovery at Scale
|
Nov 10, 2025 |
|
AdaFlow: Variance-Adaptive Flow-Based Imitation Learning
|
Nov 10, 2025 |
|
A Framework for LLM Application Safety Evaluation
|
Nov 10, 2025 |
|
SuperBPE: Space Travel for Language Models
|
Nov 04, 2025 |
|
Samsung: zFLoRA: Zero-Latency Fused Low-Rank Adapters
|
Nov 04, 2025 |
|
PolicySmith: Automated Systems Heuristic Generation via LLMs
|
Nov 04, 2025 |
|
MorphKV: Constant-Sized KV Caches for LLM Inference
|
Nov 04, 2025 |
|
HALoS: Hierarchical Asynchronous LLM Training over Slow Networks
|
Nov 04, 2025 |
|
Gumbel-Softmax for Differentiable Categorical Reparameterization and Selective Networks
|
Nov 04, 2025 |
|
Google: Supervised Reinforcement Learning for Step-wise Reasoning in LLMs
|
Nov 04, 2025 |
|
Anchored Diffusion Language Model: Superior Generation and Reasoning
|
Nov 04, 2025 |
|
RetNet: Retentive Networks: Transformer Successor for Large Language Models
|
Nov 02, 2025 |
|
Kimi Linear: Efficient Expressive Attention Architecture
|
Nov 02, 2025 |
|
ALiBi: Attention with Linear Biases Enables Length Extrapolation
|
Nov 01, 2025 |
|
Small Versus Large Models for Requirements Classification
|
Oct 31, 2025 |
|
Quest: Query-Aware Sparsity for Efficient LLM Inference
|
Oct 31, 2025 |
|
Hyper-Scaling LLM Inference with KV Cache Compression
|
Oct 31, 2025 |
|
Flash-LLM: Efficient LLM Inference with Unstructured Sparsity on Tensor Cores
|
Oct 31, 2025 |
|
ELASTIC: Linear Attention for Sequential Interest Compression
|
Oct 31, 2025 |
|
Architectural Scaling Laws for Efficient LLMs
|
Oct 31, 2025 |
|
Anthropic: Introspective Awareness in LLMs
|
Oct 31, 2025 |
|
TxGNN: Foundation Model for Zero-Shot Drug Repurposing
|
Oct 29, 2025 |
|
Sentence-BERT: Siamese Networks for Sentence Embeddings
|
Oct 29, 2025 |
|
ATTENTION2D and lean attention: Distributed Self-Attention
|
Oct 29, 2025 |
|
The Free Transformer: VAE Extension for Decoders
|
Oct 26, 2025 |
|
Survey of Emerging Topics in AI and Robotics
|
Oct 26, 2025 |
|
Stuck in the Matrix: LLM Spatial Reasoning
|
Oct 26, 2025 |
|
Structural Understanding of LLM Overthinking
|
Oct 26, 2025 |
|
Strata: Efficient Hierarchical Context Caching for LLM Serving
|
Oct 26, 2025 |
|
STAR: Sub-Entry Sharing TLB for Multi-Instance GPU Efficiency
|
Oct 26, 2025 |
|
Ring-linear: Efficient Hybrid Architecture for Long-Context Reasoning
|
Oct 26, 2025 |
|
Open-o3 Video: Spatio-Temporal Grounded Reasoning
|
Oct 26, 2025 |
|
MASA: Meta-Awareness via Self-Alignment Reinforcement Learning
|
Oct 26, 2025 |
|
Lp-Reg: Low-Probability Tokens Sustain RL Exploration
|
Oct 26, 2025 |
|
LLMs Learning from Verbal Feedback Without Scalar Rewards
|
Oct 26, 2025 |
|
LithOS: Operating System for Efficient GPU Machine Learning
|
Oct 26, 2025 |
|
LLM-Empowered Knowledge Graph Construction: A Survey
|
Oct 26, 2025 |
|
LFM2-8B-A1B: Efficient On-Device Mixture-of-Experts
|
Oct 26, 2025 |
|
Latent Constituency in Humans and LLMs
|
Oct 26, 2025 |
|
Introducing MTEB v2: Multimodal Embedding Evaluation
|
Oct 26, 2025 |
|
Internal Mechanisms of a Large Language Model
|
Oct 26, 2025 |
|
GigaBrain-0: World Model-Powered Generalist Robots
|
Oct 26, 2025 |
|
FlashAttention: IO-Aware Fast and Memory-Efficient Attention
|
Oct 26, 2025 |
|
Cognitive Impact of AI and Search on Essay Writing
|
Oct 26, 2025 |
|
Cattell–Horn–Carroll Theory of Intelligence
|
Oct 26, 2025 |
|
Structural Understanding of LLM Overthinking
|
Oct 22, 2025 |
|
RoBERTa: Robustly Optimized BERT Pretraining Approach
|
Oct 22, 2025 |
|
REFRAG: v2 paper: Efficient RAG Decoding via Context Compression
|
Oct 22, 2025 |
|
RAG-Anything: Unified Multimodal Knowledge Retrieval Framework
|
Oct 22, 2025 |
|
LLM-Guided Hierarchical Retrieval: The LATTICE Framework
|
Oct 22, 2025 |
|
LightMem: Lightweight Efficient Memory-Augmented Generation
|
Oct 22, 2025 |
|
Inheritune: Efficient LLM Training via Attention Collapse
|
Oct 22, 2025 |
|
In-Context Learning as Implicit Learning Algorithms
|
Oct 22, 2025 |
|
EssenceBench: Compressing LLM Benchmarks via Redundancy and Genetic Algorithm
|
Oct 22, 2025 |
|
Elastic-Cache: Adaptive KV Caching for Diffusion LLMs
|
Oct 22, 2025 |
|
Dr.LLM: Dynamic Layer Routing in LLMs
|
Oct 22, 2025 |
|
A Psychometric Framework for Artificial General Intelligence
|
Oct 22, 2025 |
|
Mojo: Performance-Portable HPC Kernels on GPUs
|
Oct 18, 2025 |
|
Geometric Flows of Logic in LLM Representation Space
|
Oct 18, 2025 |
|
Scaling Reinforcement Learning Compute for LLMs
|
Oct 17, 2025 |
|
HBF: High Bandwidth Flash for AI Inferencing
|
Oct 15, 2025 |
|
COPA: Composable On-Package GPU Architecture for Domain Specialization
|
Oct 15, 2025 |
|
Architectural Migration to Multi-head Latent Attention
|
Oct 15, 2025 |
|
UniVideo: Unified Video Understanding, Generation, and Editing
|
Oct 11, 2025 |
|
Training-Free GRPO: Policy Optimization via Context Space
|
Oct 11, 2025 |
|
RAND: Securing AI Model Weights: Preventing Theft and Misuse
|
Oct 11, 2025 |
|
Performance of Confidential Computing for Large Language Models
|
Oct 11, 2025 |
|
Multi-Agent Tool-Integrated Policy Optimization (MATPO)
|
Oct 11, 2025 |
|
Google: Confidential Computing with Accelerated AI Workloads on GCE
|
Oct 11, 2025 |
|
AWS: Nitro System: Security, Enclaves, and Generative AI
|
Oct 11, 2025 |
|
Anthropic: Confidential Inference via Trusted Virtual Machines
|
Oct 11, 2025 |
|
Samsung: Less is More: Recursive Reasoning with Tiny Networks
|
Oct 10, 2025 |
|
Petri: Accelerating AI Safety Auditing
|
Oct 10, 2025 |
|
Low-Precision Transformer Failure in Flash Attention
|
Oct 10, 2025 |
|
Early Experience for Language Agent Improvement
|
Oct 10, 2025 |
|
Dragon Hatchling: Brain-Inspired AI Architecture
|
Oct 10, 2025 |
|
CLUE: Hidden-State Clustering for Non-parametric Verification
|
Oct 10, 2025 |
|
Agentic Context Engineering: Evolving Contexts for Self-Improving LLMs
|
Oct 10, 2025 |
|
AGENTFLOW: In-the-Flow Agentic System Optimization
|
Oct 10, 2025 |
|
Test-Time Reinforcement Learning for LLMs
|
Oct 08, 2025 |
|
ReasoningBank: Scaling Agent Self-Evolving with Reasoning Memory
|
Oct 08, 2025 |
|
Reactive Transformer: Stateful Real-Time Language Models
|
Oct 08, 2025 |
|
RECAP: Safety Alignment via Counter-Aligned Prefilling
|
Oct 08, 2025 |
|
Paris: Decentralized Open-Weight Diffusion Model
|
Oct 08, 2025 |
|
ONNX Ecosystem, Optimization, and Deployment
|
Oct 08, 2025 |
|
NIST Evaluation of DeepSeek AI Models
|
Oct 08, 2025 |
|
MotionRAG: Retrieval-Augmented Image-to-Video Generation
|
Oct 08, 2025 |
|
LongCodeZip: Compress Long Code Context for LLMs
|
Oct 08, 2025 |
|
Imperceptible Jailbreaking Against Large Language Models
|
Oct 08, 2025 |
|
Implicit Dynamics of In-Context Learning
|
Oct 08, 2025 |
|
GNN101: Visual Learning of Graph Neural Networks
|
Oct 08, 2025 |
|
Emergent Abilities of Large Language Models
|
Oct 08, 2025 |
|
DC-VideoGen: Efficient Video Generation with Deep Compression
|
Oct 08, 2025 |
|
Contextual Blocks: Implicit Weight Updates and Federated Learning
|
Oct 08, 2025 |
|
CoDA: Collaborative Multi-Agent Data Visualization
|
Oct 08, 2025 |
|
Analog In-Memory Attention for Energy-Efficient LLMs
|
Oct 08, 2025 |
|
ACON: Optimizing Context Compression for LLM Agents
|
Oct 08, 2025 |
|
Regression Language Models for Code Metrics
|
Oct 03, 2025 |
|
Introducing RTEB: Retrieval Embedding Benchmark
|
Oct 03, 2025 |
|
CUDA Unified Memory and Heterogeneous Memory Management
|
Oct 02, 2025 |
|
Moravec's Paradox and AI Automation Limits
|
Oct 01, 2025 |
|
Characterizing LLM KV Cache Workloads in Production
|
Oct 01, 2025 |
|
BurstGPT: A Real-World LLM Serving Workload Dataset
|
Oct 01, 2025 |
|
Qwen3-Next & Qwen3-Omni technical report
|
Sep 30, 2025 |
|
Variational Reasoning Framework for Language Models
|
Sep 29, 2025 |
|
Schoenfeld Theory Applied to Large Reasoning Models
|
Sep 27, 2025 |
|
Federated Learning with Soft Embeddings for Retrieval
|
Sep 27, 2025 |
|
CWM: Code Generation with World Models
|
Sep 27, 2025 |
|
Tree-based Group Policy Optimization for LLM Agents
|
Sep 26, 2025 |
|
Seedream 4.0: Multimodal Image Generation System
|
Sep 26, 2025 |
|
GDPval: Measuring AI Performance on Real-World Work
|
Sep 26, 2025 |
|
EmbeddingGemma: Powerful Lightweight Text Representations
|
Sep 26, 2025 |
|
CE-GPPO: Controlling Entropy via Gradient-Preserving Policy Optimization
|
Sep 26, 2025 |
|
LLM-I: Interleaved Multimodal Creators via Tool-Use
|
Sep 20, 2025 |
|
Adaptive Compression Techniques for Efficient LLM Inference
|
Sep 20, 2025 |
|
THOR: Hierarchical RL for Mathematical Reasoning
|
Sep 19, 2025 |
|
The Uneven Diffusion of AI Adoption
|
Sep 19, 2025 |
|
Single-stream Policy Optimization for LLMs
|
Sep 19, 2025 |
|
SearchInstruct: Instruction Tuning with Dynamic Retrieval
|
Sep 19, 2025 |
|
FlowRL: Distribution Matching for LLM Reasoning
|
Sep 19, 2025 |
|
Evolving Language Models Without Labels: EVOL-RL
|
Sep 19, 2025 |
|
REFRAG: Rethinking RAG-based Decoding
|
Sep 18, 2025 |
|
Pre-computing & reusing KV caches to accelerate RAG inference
|
Sep 18, 2025 |
|
DeepSeek-R1: Reinforcing LLM Reasoning Through Self-Evolution
|
Sep 18, 2025 |
|
WebSailor-V2: Bridging Proprietary Agents with Synthetic Data and RL
|
Sep 17, 2025 |
|
TailorKV: Hybrid KV Cache Compression for LLMs
|
Sep 17, 2025 |
|
ShadowKV: High-Throughput Long-Context LLM Inference
|
Sep 17, 2025 |
|
QuantAgent: Multi-Agent LLM for High-Frequency Trading
|
Sep 17, 2025 |
|
MIRAGE: Optimizing LLM KV Cache with Parameter Remapping
|
Sep 17, 2025 |
|
LoFT: Parameter-Efficient Fine-Tuning for Long-tailed Semi-Supervised Learning
|
Sep 17, 2025 |
|
Infini-gram: Scaling Unbounded N-gram Language Models
|
Sep 17, 2025 |
|
Dynamic Chunking for Hierarchical Sequence Modeling
|
Sep 17, 2025 |
|
UQ: Unsolved Questions for Language Models
|
Sep 16, 2025 |
|
PETALS: Collaborative Large Language Model Inference and Fine-tuning
|
Sep 16, 2025 |
|
Non-Penetrative Tensor Partitioning for Collaborative AIoT Inference
|
Sep 16, 2025 |
|
Native Sparse Attention: Efficient Long-Context LLMs
|
Sep 16, 2025 |
|
Janus-Pro: Unified Multimodal AI with Scaled Improvements
|
Sep 16, 2025 |
|
Hierarchical Reasoning Model: Brain-Inspired AI for Complex Tasks
|
Sep 16, 2025 |
|
Generalist Reward Modeling with Inference-Time Scaling
|
Sep 16, 2025 |
|
Federated Post-Training LLMs: An Accessibility and Efficiency Survey
|
Sep 16, 2025 |
|
Collaborative Edge Inference with Dynamic Task Offloading and Early Exiting
|
Sep 16, 2025 |
|
CodeI/O: Reasoning Patterns Through Code Input-Output Prediction
|
Sep 16, 2025 |
|
Adaptive LLM Partitioning for Edge Inference
|
Sep 16, 2025 |
|
The Illusion of Diminishing Returns in LLM Execution
|
Sep 15, 2025 |
|
PyTorch FSDP: Scaling Fully Sharded Data Parallel
|
Sep 15, 2025 |
|
MetaGraph: knowledge graphs from financial NLP
|
Sep 15, 2025 |
|
Hallucination to Truth: A Review of Fact-Checking and Factuality
Evaluation in Large Language Model
|
Sep 15, 2025 |
|
HybridServe: Efficient LLM Inference with Hybrid Caching
|
Sep 15, 2025 |
|
GraphSAGE: Inductive Representation Learning on Large Graphs
|
Sep 15, 2025 |
|
FlexGen: High-Throughput LLM Inference on a Single GPU
|
Sep 15, 2025 |
|
AWQ: On-Device LLM Compression and Acceleration
|
Sep 15, 2025 |
|
Llama 3: Architecture, Capabilities, and Safety
|
Sep 14, 2025 |
|
Graph Patterns of Knowledge in Large Language Models
|
Sep 14, 2025 |
|
Survey of Reinforcement Learning for Large
Reasoning Models
|
Sep 13, 2025 |
|
Statistical Methods for Generative AI Reliability
|
Sep 13, 2025 |
|
SpikingBrain: Brain-Inspired LLMs for Efficient Long-Context Processing
|
Sep 13, 2025 |
|
EntiGraph: Scaling Language Models with Synthetic Pretraining
|
Sep 13, 2025 |
|
All for One: LLMs Solve Mental Math at the Last Token
|
Sep 13, 2025 |
|
Parallel-R1: Reinforcement Learning for Parallel Thinking in LLMs
|
Sep 12, 2025 |
|
NOVELTYBENCH: Evaluating Language Model Diversity
|
Sep 12, 2025 |
|
HyperController: Fast, Stable Reinforcement Learning Hyperparameter Optimization
|
Sep 12, 2025 |
|
Explaining AI for Digital Advertising with LLMs
|
Sep 11, 2025 |
|
ByteCheckpoint: A Unified LLM Checkpointing System
|
Sep 11, 2025 |
|
AdLlama: Boosting Ad Performance with Reinforcement Learning
|
Sep 11, 2025 |
|
Mini-o3: Scaling Reasoning for Visual Search
|
Sep 10, 2025 |
|
Masked Diffusion Models: Performance and Theory
|
Sep 10, 2025 |
|
K2-Think: A Parameter-Efficient Reasoning System
|
Sep 10, 2025 |
|
INF2: Near-Storage LLM Inference for High Throughput
|
Sep 10, 2025 |
|
Darling: Reinforcing Diversity and Quality in Language Models
|
Sep 10, 2025 |
|
BLEU: Automatic Machine Translation Evaluation
|
Sep 10, 2025 |
|
AlphaEvolve: AI for Scientific and Algorithmic Discovery
|
Sep 10, 2025 |
|
TraceRL: Reinforcement Learning for Diffusion Language Models
|
Sep 09, 2025 |
|
LLM Benchmark Robustness to Linguistic Variation
|
Sep 09, 2025 |
|
Behavioral Fingerprinting of Large Language Models
|
Sep 09, 2025 |
|
Offloading LLM Models and KV Caches to NVMe SSDs
|
Sep 08, 2025 |
|
SGLang: Efficient Language Model Program Execution
|
Sep 07, 2025 |
|
OpenELM: Apple's Open Language Model Family
|
Sep 07, 2025 |
|
GPT-NeoX: Large-Scale Autoregressive Language Modeling in PyTorch
|
Sep 07, 2025 |
|
FineVision: Open Data for Computer Vision
|
Sep 07, 2025 |
|
Evaluating Large Language Models Trained on Code
|
Sep 07, 2025 |
|
Eleuther: evaluating LLMs
|
Sep 07, 2025 |
|
Democratizing AI Compute: The Modular Vision
|
Sep 07, 2025 |
|
SAIR: Accelerating Pharma R&D with AI-Powered Structural Intelligence
|
Sep 06, 2025 |
|
Limitations of Embedding-Based Retrieval
|
Sep 06, 2025 |
|
MTEB & MMTEB: The Massive Text Embedding Benchmark
|
Sep 05, 2025 |
|
Inverse IFEval: Unlearning LLM Cognitive Inertia
|
Sep 05, 2025 |
|
EmbeddingGemma: On-Device AI for High-Quality Embeddings
|
Sep 05, 2025 |
|
DeepResearch Arena: Benchmarking LLMs' Research Abilities
|
Sep 05, 2025 |
|
The Rise of Physical Neural Networks
|
Sep 04, 2025 |
|
Supervised Learning in DNA Neural Networks
|
Sep 04, 2025 |
|
FastVLM: Efficient Vision Encoding for Language Models
|
Sep 04, 2025 |
|
Apertus Tech Report Overview
|
Sep 04, 2025 |
|
Scientific LLMs: A Data-Centric Survey and Roadmap
|
Sep 03, 2025 |
|
rStar2-Agent: Smarter Math Reasoning Through Agentic RL
|
Sep 03, 2025 |
|
FusionANNS: Billion-Scale ANNS with SSD and GPU
|
Sep 03, 2025 |
|
Training Recurrent Neural Networks: Vanishing and Exploding Gradients
|
Aug 27, 2025 |
|
SPAM: Stabilizing LLM Training with Spike-Aware Optimization
|
Aug 27, 2025 |
|
Pimba: Processing-in-Memory for LLM Serving
|
Aug 27, 2025 |
|
Oaken: Fast, Efficient LLM Serving with Hybrid KV Cache Quantization
|
Aug 27, 2025 |
|
AdamW: Decoupled Weight Decay Regularization for Adaptive Gradient Algorithms
|
Aug 27, 2025 |
|
Adafactor: Memory-Efficient Adaptive Learning Rates
|
Aug 27, 2025 |
|
Quantizing Diffusion LLMs: A Systematic Study
|
Aug 26, 2025 |
|
ODYSSEY: Unified Mobile Manipulation for Agile Quadruped Robots
|
Aug 26, 2025 |
|
Google: Measuring AI's Environmental Impact at Scale
|
Aug 26, 2025 |
|
ComoRAG: Cognitively Inspired Narrative Reasoning
|
Aug 26, 2025 |
|
GPT-5 Spatial Intelligence: An Empirical Study
|
Aug 24, 2025 |
|
DeepSeek-V3.1: A Hybrid AI Model with Enhanced Reasoning
|
Aug 23, 2025 |
|
Compressed Experts: Efficient MoE Model Editing
|
Aug 23, 2025 |
|
Genie 3: A New Frontier for World Models
|
Aug 22, 2025 |
|
Los Alamos: overcoming the memory wall fighting sparse memory access
|
Aug 21, 2025 |
|
Switch Transformers: Trillion Parameter Models with Sparsity
|
Aug 20, 2025 |
|
Speed Always Wins: Efficient Large Language Model Architectures
|
Aug 20, 2025 |
|
Linear Transformers: Faster Than RNNs
|
Aug 20, 2025 |
|
Self-Search Reinforcement Learning for LLMs
|
Aug 18, 2025 |
|
Diffusion Language Models: Principles, Techniques, and Applications
|
Aug 18, 2025 |
|
Continuous Batching for LLM Inference: Throughput and Latency Gains
|
Aug 18, 2025 |
|
Atom: Low-Bit Quantization for LLM Serving
|
Aug 18, 2025 |
|
The Mapped Memory Mistake: Why DBMSs Should Avoid MMAP
|
Aug 13, 2025 |
|
Scaling PostgreSQL at OpenAI: Read-Heavy Workloads and Optimizations - PGConf.dev 2025
|
Aug 13, 2025 |
|
pNFS Flex Files
|
Aug 13, 2025 |
|
NVMe Offload on Colossal AI: Breaking the GPU Memory Wall
|
Aug 13, 2025 |
|
NVIDIA GDS, BAM Vs RocM solutions
|
Aug 13, 2025 |
|
fMoE: Fine-Grained Expert Offloading for MoE Serving
|
Aug 13, 2025 |
|
ELMo-Tune-V2: LLM-Assisted Auto-Tuning for Key-Value Stores
|
Aug 13, 2025 |
|
AiSAQ: DRAM-free ANNS with Product Quantization
|
Aug 13, 2025 |
|
Qwen-Image: Generation and Editing with Precision
|
Aug 12, 2025 |
|
Mem0: Scalable Long-Term Memory for AI Agents
|
Aug 12, 2025 |
|
Chain-of-Thought Reasoning: A Brittle Mirage?
|
Aug 11, 2025 |
|
Xavier Initialization: Deep Feedforward Networks: Training Difficulties and Solutions
|
Aug 08, 2025 |
|
ZeRO-Offload: Democratizing Billion-Scale Model Training
|
Aug 08, 2025 |
|
vLLM & PagedAttention: Efficient LLM Serving with Virtual Memory
|
Aug 08, 2025 |
|
TierTrain: Proactive Memory Tiering for DNN Training
|
Aug 08, 2025 |
|
Test-Time Scaling
|
Aug 08, 2025 |
|
The Elements of Differentiable Programming
|
Aug 08, 2025 |
|
Teraio: Cost-Efficient LLM Training via Lifetime-Aware Tensor Offloading
|
Aug 08, 2025 |
|
MoE Offloaded
|
Aug 08, 2025 |
|
Shared Virtual Memory: Design and Performance Implications
|
Aug 08, 2025 |
|
Reinforcement Learning
|
Aug 08, 2025 |
|
Reinforcement Pre-Training for Language Models
|
Aug 08, 2025 |
|
PartitionedVC: Optimizing External Memory Graph Analytics
|
Aug 08, 2025 |
|
Multiagent Debate Improves Language Model Reasoning
|
Aug 08, 2025 |
|
Multi Query Attention: PaLM: Scaling Language Modeling with Pathways
|
Aug 08, 2025 |
|
MUVERA: Efficient Multi-Vector Information Retrieval
|
Aug 08, 2025 |
|
Mu: The Windows Settings Agent Language Model
|
Aug 08, 2025 |
|
Movement Pruning: Adaptive Sparsity by Fine-Tuning
|
Aug 08, 2025 |
|
Mistral 7B: Superior Performance in a Smaller Package
|
Aug 08, 2025 |
|
MetaScale: Test-Time Scaling with Evolving Meta-Thoughts
|
Aug 08, 2025 |
|
Microscaling Quantization for Large Language Models
|
Aug 08, 2025 |
|
Lost in the Middle: How Language Models Use Long Contexts
|
Aug 08, 2025 |
|
MEGABYTE: Multiscale Transformers for Million-byte Sequences
|
Aug 08, 2025 |
|
Mamba: Linear-Time Sequence Modeling with Selective State Spaces
|
Aug 08, 2025 |
|
LoRA: Low-Rank Adaptation of Large Language Models
|
Aug 08, 2025 |
|
LMCache: Supercharging LLM Performance with KV Cache Management
|
Aug 08, 2025 |
|
KVQuant: LLM Inference with KV Cache Quantization
|
Aug 08, 2025 |
|
Kaiming Initialization and PReLU
|
Aug 08, 2025 |
|
Gemma: Google DeepMind's Open Language Models
|
Aug 08, 2025 |
|
Fast Learning for Deep Belief Networks
|
Aug 08, 2025 |
|
FP8 Quantization
|
Aug 08, 2025 |
|
FlashAttention-2: Faster Attention with Better Parallelism
|
Aug 08, 2025 |
|
DyNN-Offload: Efficient Memory for Dynamic Neural Networks
|
Aug 08, 2025 |
|
Dynamic Tanh: Transformers Without Normalization
|
Aug 08, 2025 |
|
DroidSpeak: Cross-LLM KV Cache Sharing
|
Aug 08, 2025 |
|
DiMSUM: Image Generation with Diffusion Mamba
|
Aug 08, 2025 |
|
DeepSeek Safety Concerns
|
Aug 08, 2025 |
|
Diffusion Probabilitic Models: Deep Unsupervised Learning Through Diffusion Processes
|
Aug 08, 2025 |
|
Demystifying Mamba: Architecture and Capabilities
|
Aug 08, 2025 |
|
DeepSeek-R1: Incentivizing Reasoning in LLMs
|
Aug 08, 2025 |
|
DeepSeek-V3: A Technical Report
|
Aug 08, 2025 |
|
DeepSeekMoE: Scalable Mixture-of-Experts Language Models
|
Aug 08, 2025 |
|
DeepSeek-R1 Dynamic 1.58-bit Quantization: A Performance Analysis
|
Aug 08, 2025 |
|
Concept Drift
|
Aug 08, 2025 |
|
ColBERT and ColBERT v2
|
Aug 08, 2025 |
|
CODEGEN: Open Language Model for Code Synthesis
|
Aug 08, 2025 |
|
Chain of thought
|
Aug 08, 2025 |
|
AI and the Memory Wall: Overcoming Bottlenecks
|
Aug 08, 2025 |
|
AdamW: Decoupled Weight Decay Regularization for Adaptive Gradient Algorithms
|
Aug 08, 2025 |
|
Adam: A Method for Stochastic Optimization
|
Aug 08, 2025 |
|
Transformer Scaling
|
Aug 07, 2025 |
|
Scaling Laws
|
Aug 07, 2025 |
|
ResNets - residual block
|
Aug 07, 2025 |
|
RoPE
|
Aug 07, 2025 |
|
LSTM: the forget gate
|
Aug 07, 2025 |
|
Longformer: A Transformer for Long Documents
|
Aug 07, 2025 |
|
Linformer: Efficient Self-Attention with Linear Complexity
|
Aug 07, 2025 |
|
LeNet-5: Convolutional Networks for Character Recognition
|
Aug 07, 2025 |
|
Layer Normalization and Dual Patch Normalization
|
Aug 07, 2025 |
|
Learning from repeated data
|
Aug 07, 2025 |
|
GPT4 Technical Report
|
Aug 07, 2025 |
|
GQA: Grouped Query Attention
|
Aug 07, 2025 |
|
GPT3
|
Aug 07, 2025 |
|
GPT2
|
Aug 07, 2025 |
|
GELU
|
Aug 07, 2025 |
|
Dropout
|
Aug 07, 2025 |
|
Constitutional AI: Harmlessness Through Self-Improvement
|
Aug 07, 2025 |
|
Chinchilla: Optimal Language Model Scaling
|
Aug 07, 2025 |
|
Chamfer Matching: Image Registration and Medical Applications
|
Aug 07, 2025 |
|
BART
|
Aug 07, 2025 |
|
BERT
|
Aug 07, 2025 |
|
Batch Normalization
|
Aug 07, 2025 |
|
Attention is all you need
|
Aug 07, 2025 |