7 papers
Training LLMs for Multi-Step Tool Orchestration with Constrained Data Synthesis and Graduated Rewards
Cheng Jiayang, Xin Liu, Zhihan Zhang +8
Multi-step tool orchestration remains challenging for LLMs, as state-of-the-art models frequently fail on full sequence execution due to parameter errors. Training for these workfl…
Intern-S1-Pro: Scientific Multimodal Foundation Model at Trillion Scale
Yicheng Zou, Dongsheng Zhu, Lin Zhu +174
We introduce Intern-S1-Pro, the first one-trillion-parameter scientific multimodal foundation model. Scaling to this unprecedented size, the model delivers a comprehensive enhancem…
Does Explanation Correctness Matter? Linking Computational XAI Evaluation to Human Understanding
Gregor Baer, Chao Zhang, Isel Grau +1
Explainable AI (XAI) methods are commonly evaluated using functional correctness metrics, sometimes termed faithfulness or fidelity, which estimate how closely an explanation refle…
Self-Rewarding PPO: Aligning Large Language Models with Demonstrations Only
Qingru Zhang, Liang Qiu, Ilgee Hong +11
Supervised fine-tuning (SFT) has emerged as a crucial method for aligning large language models (LLMs) with human-annotated demonstrations. However, SFT, being an off-policy approa…
WebAgent-R1: Training Web Agents via End-to-End Multi-Turn Reinforcement Learning
Zhepei Wei, Wenlin Yao, Yao Liu +9
While reinforcement learning (RL) has demonstrated remarkable success in enhancing large language models (LLMs), it has primarily focused on single-turn tasks such as solving math…
AutoMixAlign: Adaptive Data Mixing for Multi-Task Preference Optimization in LLMs
Nicholas E. Corrado, Julian Katz-Samuels, Adithya Devraj +6
When aligning large language models (LLMs), their performance on various tasks (such as being helpful, harmless, and honest) depends heavily on the composition of their training da…