When Does Trajectory-Level Supervision Permit Efficient Offline Reinforcement Learning?
Paper • 2606.18531 • Published • 3 •
2
Noah
noahml
AI & ML interests
None yet
Recent Activity
commentedon a paper about 3 hours ago
When Does Trajectory-Level Supervision Permit Efficient Offline Reinforcement Learning? commentedon a paper about 3 hours ago
SciOrch: Learning to Orchestrate Expert LLMs for Solving Frontier Multimodal Scientific Reasoning Tasks commentedon a paper about 3 hours ago
FlowBender: Feedback-Aware Training for Self-Correcting Conditional FlowsOrganizations
None yet
commented 5
papers about 3 hours ago
SciOrch: Learning to Orchestrate Expert LLMs for Solving Frontier Multimodal Scientific Reasoning Tasks
Paper • 2606.15872 • Published • 4 •
3
FlowBender: Feedback-Aware Training for Self-Correcting Conditional Flows
Paper • 2606.20404 • Published • 12 •
2
Multi-LCB: Extending LiveCodeBench to Multiple Programming Languages
Paper • 2606.20517 • Published • 5 •
2
DragMesh-2: Physically Plausible Dexterous Hand-Object Interaction with Articulated Objects
Paper • 2606.15133 • Published • 45 •
3
commented 5
papers about 21 hours ago
STARE: Surprisal-Guided Token-Level Advantage Reweighting for Policy Entropy Stability
Paper • 2606.19236 • Published • 8 •
4
A Benchmark and Framework for Evaluating Next Action Predictions in Spreadsheets
Paper • 2606.13802 • Published •
3
LLM-Enabled NWDAF: A Step Toward AI-Native 6G Network Intelligence
Paper • 2606.11877 • Published •
3
Morpheus: A Morphology-Aware Neural Tokenizer and Word Embedder for Turkish
Paper • 2606.18717 • Published • 1 •
2
From Trainee to Trainer: LLM-Designed Training Environment for RL with Multi-Agent Reasoning
Paper • 2606.17682 • Published • 16 •
2
commented 5
papers 1 day ago
Variable-Width Transformers
Paper • 2606.18246 • Published • 6 •
1
Visual-Seeker: Towards Visual-Native Multimodal Agentic Search via Active Visual Reasoning
Paper • 2606.15231 • Published • 3 •
3
Guava: An Effective and Universal Harness for Embodied Manipulation
Paper • 2606.18363 • Published • 24 •
5
EfficientRollout: System-Aware Self-Speculative Decoding for RL Rollouts
Paper • 2606.18967 • Published • 20 •
3
SAE Interventions are Unreliable: Post-Intervention Recovery of Suppressed Behavior
Paper • 2606.18322 • Published • 16 •
3
commented 5
papers 2 days ago
GameCraft-Bench: Can Agents Build Playable Games End-to-End in a Real Game Engine?
Paper • 2606.17861 • Published • 44 •
5
Learning from the Self-future: On-policy Self-distillation for dLLMs
Paper • 2606.18195 • Published • 70 •
2
Rethinking the Role of Efficient Attention in Hybrid Architectures
Paper • 2606.15378 • Published • 14 •
3
Show the Signal, Hide the Noise: Spectral Forcing for Pixel-Space Diffusion
Paper • 2606.15236 • Published • 18 •
3
ACE-Ego-0: Unifying Egocentric Human and Robotic Data for VLA Pretraining
Paper • 2606.17200 • Published • 43 •
3