11 papers
CAS: Conformalized Agentic Search via Adaptive Retrieval and Policy Weighting
Zixi Zhu, Jiayuan Su, Jian Zhang +2
Search Agents face a severe reliability crisis during reinforcement learning (RL) fine-tuning. Heuristic Top-K retrieval often causes critical evidence loss or noise inclusion, whi…
SheetBrain: A Neuro-Symbolic Agent for Accurate Reasoning over Complex and Large Spreadsheets
Ziwei Wang, Jiayuan Su, Mengyu Zhou +7
Understanding and reasoning over complex spreadsheets remain fundamental challenges for large language models (LLMs), which often struggle with accurately capturing the complex str…
CP-Router: An Uncertainty-Aware Router Between LLM and LRM
Jiayuan Su, Fulin Lin, Zhaopeng Feng +7
Recent advances in Large Reasoning Models (LRMs) have significantly improved long-chain reasoning capabilities over Large Language Models (LLMs). However, LRMs often produce unnece…
MT: Scaling MLLM-based Text Image Machine Translation via Multi-Task Reinforcement Learning
Zhaopeng Feng, Yupu Liang, Shaosheng Cao +7
Text Image Machine Translation (TIMT)-the task of translating textual content embedded in images-is critical for applications in accessibility, cross-lingual information access, an…
MT-R1-Zero: Advancing LLM-based Machine Translation via R1-Zero-like Reinforcement Learning
Zhaopeng Feng, Shaosheng Cao, Jiahan Ren +7
Large-scale reinforcement learning (RL) methods have proven highly effective in enhancing the reasoning abilities of large language models (LLMs), particularly for tasks with verif…
MT-RewardTree: A Comprehensive Framework for Advancing LLM-Based Machine Translation via Reward Modeling
Zhaopeng Feng, Jiahan Ren, Jiayuan Su +3
Process reward models (PRMs) have shown success in complex reasoning tasks for large language models (LLMs). However, their application to machine translation (MT) remains underexp…
Class Incremental Fault Diagnosis under Limited Fault Data via Supervised Contrastive Knowledge Distillation
Hanrong Zhang, Yifei Yao, Zixuan Wang +4
Class-incremental fault diagnosis requires a model to adapt to new fault classes while retaining previous knowledge. However, limited research exists for imbalanced and long-tailed…
M-MAD: Multidimensional Multi-Agent Debate for Advanced Machine Translation Evaluation
Zhaopeng Feng, Jiayuan Su, Jiamei Zheng +5
Recent advancements in large language models (LLMs) have given rise to the LLM-as-a-judge paradigm, showcasing their potential to deliver human-like judgments. However, in the fiel…
PersonaMath: Boosting Mathematical Reasoning via Persona-Driven Data Augmentation
Jing Luo, Longze Chen, Run Luo +12
While closed-source Large Language Models (LLMs) demonstrate strong mathematical problem-solving abilities, open-source models still face challenges with such tasks. To bridge this…
API Is Enough: Conformal Prediction for Large Language Models Without Logit-Access
Jiayuan Su, Jing Luo, Hongwei Wang +1
This study aims to address the pervasive challenge of quantifying uncertainty in large language models (LLMs) without logit-access. Conformal Prediction (CP), known for its model-a…
Dialectical Alignment: Resolving the Tension of 3H and Security Threats of LLMs
Shu Yang, Jiayuan Su, Han Jiang +5
With the rise of large language models (LLMs), ensuring they embody the principles of being helpful, honest, and harmless (3H), known as Human Alignment, becomes crucial. While exi…