6 papers
CRISP: Compressed Reasoning via Iterative Self-Policy Distillation
Hejian Sang, Yuanda Xu, Zhengze Zhou +3
Reasoning models often generate far more tokens than a task requires, which raises inference cost and can compound errors. We introduce CRISP (Compressed Reasoning via Iterative Se…
OmniUMI: Towards Physically Grounded Robot Learning via Human-Aligned Multimodal Interaction
Shaqi Luo, Yuanyuan Li, Youhao Hu +7
UMI-style interfaces enable scalable robot learning, but existing systems remain largely visuomotor, relying primarily on RGB observations and trajectory while providing only limit…
PACED: Distillation and On-Policy Self-Distillation at the Frontier of Student Competence
Yuanda Xu, Hejian Sang, Zhengze Zhou +2
Standard LLM distillation treats all training problems equally -- wasting compute on problems the student has already mastered or cannot yet solve. We empirically show that this in…
Not all tokens are needed(NAT): token efficient reinforcement learning
Hejian Sang, Yuanda Xu, Zhengze Zhou +2
Reinforcement learning (RL) has become a key driver of progress in large language models, but scaling RL to long chain-of-thought (CoT) trajectories is increasingly constrained by…
Semantic Search At LinkedIn
Fedor Borisyuk, Sriram Vasudevan, Muchen Wu +71
Semantic search with large language models (LLMs) enables retrieval by meaning rather than keyword overlap, but scaling it requires major inference efficiency advances. We present…
MixLM: High-Throughput and Effective LLM Ranking via Text-Embedding Mix-Interaction
Guoyao Li, Ran He, Shusen Jing +21
Large language models (LLMs) excel at capturing semantic nuances and therefore show impressive relevance ranking performance in modern recommendation and search systems. However, t…