On the Generalization Limitation of On-Policy Distillation for Post-RL Consolidation
Paper • 2611.05609 • Published • 24
Jenny Waldman
j-waldman
·
AI & ML interests
ColBERT-style retrieval, embedding fine-tuning, rerankers, RAG, open-domain QA, diffusion policies
Recent Activity
Organizations
None yet
upvoted 4
papers 2 days ago
PyRUA-DataFarm: LLM Agent as a Cost-Efficient Teleoperator for Real-Ready Robot Data
Paper • 2611.05320 • Published • 27
MUON-RT Reusing the Teacher's Trajectory for Momentum-Free Muon for LLM Self-Distillation
Paper • 2610.41104 • Published • 23
Human-to-Robot Transfer via Bridging Representation and Action Space in VLAs
Paper • 2611.04391 • Published • 16
liked a
model 3 days ago
upvoted 3
papers 3 days ago
RoboColosseum: Evaluating Physical Intelligence at Scale In-The-Wild
Paper • 2611.02497 • Published • 51
Words Caught in the Heart: Simulating Person-Specific Expression for Self-Opaque Clients in Psychotherapy
Paper • 2611.00617 • Published • 78
Do We Need an LLM to Know Why Sources Conflict? Decision Models for Knowledge-Conflict Typing in RAG
Paper • 2610.36915 • Published • 6
upvoted 4
papers 4 days ago
DeFT-RAG: Decomposing Multi-Hop Questions into Self-Contained Sub-Questions with Reusable Answers
Paper • 2610.37826 • Published • 15
CUA-Master: Scaling Computer-use Agents through Open-World Recursive Improvements
Paper • 2611.03126 • Published • 47
Lost in Recovery: Demystifying Recovery Behavior of Computer-Use Agents via Goal-Oriented Action Replay
Paper • 2610.27954 • Published • 36
PhotoGuider: Composition-Aware Cropping and 360^{circ} View Framing with Explainable Photographic Assessments
Paper • 2611.02897 • Published • 39
upvoted 4
papers 5 days ago
Wizard: Manipulate Any World in Real Time via Agentic Control
Paper • 2610.40339 • Published • 27
RoboArchitect: Automating the Design of Robotic Agents with Persistent Memory
Paper • 2610.42141 • Published • 9
TernScope: Model-Hardware Co-Design for Efficient Ternary Endoscopic Segmentation
Paper • 2610.29617 • Published • 4
MemSteer: Turning User Histories into Supervision for Agent Memory Optimization
Paper • 2610.39276 • Published • 10
liked 2
models 6 days ago
upvoted 2
papers 6 days ago