PN-Mix: Pair-Norm Guided Mixed-Precision Key Cache Quantization for Efficient Long-Output LLM Inference
Paper • 2610.19605 • Published • 6
MBerger
SHakeShakeShake
AI & ML interests
None yet
Recent Activity
upvoted a paper 3 days ago
SpecMRL: Redesign Matryoshka Representation Learning within Parametric SpectrumOrganizations
None yet
upvoted 5
papers 3 days ago
Terminal-RSI: Recursive Self-Improving Terminal Agents via Recursive Self-Rewrite and Policy-Refine
Paper • 2610.22192 • Published • 41
SpecMRL: Redesign Matryoshka Representation Learning within Parametric Spectrum
Paper • 2610.26824 • Published • 87
Cortex: Measurement-Gated Evolution and Pruning of Coding-Agent Skills and Rules
Paper • 2610.14216 • Published • 5
What Really Works for Universal Multimodal Embeddings? A Controlled Study of Contrastive Training Strategies
Paper • 2610.15035 • Published • 15
upvoted 4
papers 23 days ago
Understanding On-Policy Distillation: A Mechanistic Interpretability Perspective via Sparse Crosscoders
Paper • 2609.35210 • Published • 7
Fractional State Space Transition for Long Sequence Modeling
Paper • 2609.36314 • Published • 11
Real2Gym: Building Gyms from Videos, Bringing Skills to Robots
Paper • 2609.37089 • Published • 8
Periodic Weak Spots: Phase Sensitivity from Chunked KV-Cache Compression
Paper • 2609.36322 • Published • 120
upvoted 11
papers 4 months ago
Embodied-R1.5: Evolving Physical Intelligence via Embodied Foundation Models
Paper • 2606.11324 • Published • 166
Verifiable Environments Are LEGO Bricks: Recursive Composition for Reasoning Generalization
Paper • 2606.12373 • Published • 8
EvoTrainer: Co-Evolving LLM Policies and Training Harnesses for Autonomous Agentic Reinforcement Learning
Paper • 2606.03108 • Published • 13
Breaking Entropy Bounds: Accelerating RL Training via MTP with Rejection Sampling
Paper • 2606.12370 • Published • 22
ICA Lens: Interpreting Language Models Without Training Another Dictionary
Paper • 2606.11722 • Published • 18
InternVideo3: Agentify Foundation Models with Multimodal Contextual Reasoning
Paper • 2606.12195 • Published • 25
ComBench: A Benchmark for Rigorous Proof Reasoning and Constructive Realization in Olympiad-Level Combinatorics
Paper • 2606.10479 • Published • 21
World Pilot: Steering Vision-Language-Action Models with World-Action Priors
Paper • 2606.12403 • Published • 27
Reason, Then Re-reason: Cross-view Revisiting Improves Spatial Reasoning
Paper • 2606.11683 • Published • 31
DeNovoSWE: Scaling Long-Horizon Environments for Generating Entire Repositories from Scratch
Paper • 2606.10728 • Published • 38
Beyond Scalar Rewards by Internalizing Reasoning into Score Distributions
Paper • 2606.09076 • Published • 68