1 citations · 1 across the 14 of their papers we have counts for
15 papers
Ask, Relax, or Act? Evaluating Actionable Indeterminacy in LLM Preference Reasoning
Ang Li, Yue Lin, Feifei Kou +6
An LLM agent can recognize uncertainty yet still choose the wrong next step: asking when action is already justified, or seeking clarification when the constraints must change. We…
TAPO: Transition-Aware Policy Optimization for LLM Agents
Cong Li, Peixi Peng, Yisen Zhao +4
Recently, Reinforcement Learning (RL) has emerged as a crucial paradigm for the post-training of Large Language Model (LLM) agents. However, existing methods predominantly rely on…
Search-on-Graph-R1: Training Large Language Models to Search Knowledge Graphs with Reinforcement Learning
Jia Ao Sun, Hao Yu, Fengran Mo +4
Knowledge graph question answering (KGQA) requires navigating from topic entities to an answer several relations away. Recent methods prompt a frontier LLM to explore the graph thr…
Learning to Select, Not Relearn: Hard-Routed Mixtures of Reasoning LoRAs
Seyed Alireza Molavi, Zhan Su, Yan Hu +3
Composing independently trained LoRA adapters into a single large language model is useful for multi-domain adaptation, especially when the original training data cannot be shared.…
STOP: Structured On-Policy Pruning of Long-Form Reasoning in Low-Data Regimes
Chenjun Xu, Zhennan Zhou, Zhan Su +3
Long chain-of-thought (Long CoT) reasoning improves performance on multi-step problems, but it also induces overthinking. This inefficiency is especially problematic in low-data fi…
Efficient Data Selection for Multimodal Models via Incremental Optimization Utility
Jinhao Jing, Qiannian Zhao, Chao Huang +1
The scaling of Large Multimodal Models (LMMs) is constrained by the quality-quantity trade-off inherent in synthetic data. Previous approaches, such as LLM-as-a-Judge, have proven…
Towards Dynamic Dense Retrieval with Routing Strategy
Zhan Su, Fengran Mo, Jinghan Zhang +4
The \textit{de facto} paradigm for applying dense retrieval (DR) to new tasks involves fine-tuning a pre-trained model for a specific task. However, this paradigm has two significa…
OpenDecoder: Open Large Language Model Decoding to Incorporate Document Quality in RAG
Fengran Mo, Zhan Su, Yuchen Hui +6
The development of large language models (LLMs) has achieved superior performance in a range of downstream tasks, including LLM-based retrieval-augmented generation (RAG). The qual…
COVR:Collaborative Optimization of VLMs and RL Agent for Visual-Based Control
Canming Xia, Peixi Peng, Guang Tan +4
Visual reinforcement learning (RL) suffers from poor sample efficiency due to high-dimensional observations in complex tasks. While existing works have shown that vision-language m…
Parametric Retrieval-Augmented Generation using Latent Routing of LoRA Adapters
Zhan Su, Fengran Mo, Jinghan Zhang +3
Parametric Retrieval-Augmented Generation (PRAG) is a RAG approach that integrates external knowledge directly into model parameters using a LoRA adapter, aiming at reducing the in…
It Takes Two: Your GRPO Is Secretly DPO
Yihong Wu, Liheng Ma, Lei Ding +9
GRPO has emerged as a prominent reinforcement learning algorithm for post-training LLMs. Unlike critic-based methods, GRPO computes advantages by estimating the \emph{value baselin…
Adaptive Personalized Conversational Information Retrieval
Fengran Mo, Yuchen Hui, Yuxing Tian +5
Personalized conversational information retrieval (CIR) systems aim to satisfy users' complex information needs through multi-turn interactions by considering user profiles. Howeve…
Tensorized Clustered LoRA Merging for Multi-Task Interference
Zhan Su, Fengran Mo, Guojun Liang +4
Despite the success of the monolithic dense paradigm of large language models (LLMs), the LoRA adapters offer an efficient solution by fine-tuning small task-specific modules and m…
Smooth Reading: Bridging the Gap of Recurrent LLM to Self-Attention LLM on Long-Context Tasks
Kai Liu, Zhan Su, Peijie Dong +4
Recently, recurrent large language models (Recurrent LLMs) with linear computational complexity have re-emerged as efficient alternatives to self-attention-based LLMs (Self-Attenti…
Language Modeling Using Tensor Trains
Zhan Su, Yuqin Zhou, Fengran Mo +1
We propose a novel tensor network language model based on the simplest tensor network (i.e., tensor trains), called `Tensor Train Language Model' (TTLM). TTLM represents sentences…