5 papers
Solving Every Step Is Not Enough: Milestone Oracles Reveal a Composition Gap in LLM Math Reasoning
Zhuohan Wang, Haoran Ma, Tianyu Wu +3
Large language models (LLMs) can solve every intermediate step of a multi-step math problem on its own and still fail the full problem, even when given a roadmap of the steps and a…
Learning from Demonstrations via Capability-Aware Goal Sampling
Yuanlin Duan, Yuning Wang, Wenjie Qiu +1
Despite its promise, imitation learning often fails in long-horizon environments where perfect replication of demonstrations is unrealistic and small errors can accumulate catastro…
Learning World Models for Unconstrained Goal Navigation
Yuanlin Duan, Wensen Mao, He Zhu
Learning world models offers a promising avenue for goal-conditioned reinforcement learning with sparse rewards. By allowing agents to plan actions or exploratory goals without dir…
Exploring the Edges of Latent State Clusters for Goal-Conditioned Reinforcement Learning
Yuanlin Duan, Guofeng Cui, He Zhu
Exploring unknown environments efficiently is a fundamental challenge in unsupervised goal-conditioned reinforcement learning. While selecting exploratory goals at the frontier of…
MoE-I: Compressing Mixture of Experts Models through Inter-Expert Pruning and Intra-Expert Low-Rank Decomposition
Cheng Yang, Yang Sui, Jinqi Xiao +7
The emergence of Mixture of Experts (MoE) LLMs has significantly advanced the development of language models. Compared to traditional LLMs, MoE LLMs outperform traditional LLMs by…