Guangzhi Xiong

2 related articles

Subscribe to updates

Nature Communications Oct 28, 2026
Reward the search, not just the answer: process supervision lifts retrieval recall and holds up out of domain
Search agents for multi-hop information seeking are trained on step-level judgments of each search action instead of only on final-answer correctness. RAG-Gym casts search as a Markov decision process and compares architecture, parameter optimization and action evaluation, combining a reasoning-reflection architecture, preference optimization and critic-guided inference into Re2Search++. Trained on HotpotQA, it beats outcome-supervised baselines on the unseen Bamboogle benchmark with higher retrieval recall at comparable or fewer search steps, and a critic trained on Llama-3.1-8B raises GPT-4o-mini MedQA accuracy.
Search agents for multi-hop information seeking are trained on step-level judgments of each search action instead of only on final-answer correctness. RAG-Gym casts search as a Markov decision process and compares architecture, parameter optimization and action evaluation, combining a reasoning-reflection architecture, preference optimization and critic-guided inference into Re2Search++. Trained on HotpotQA, it beats outcome-supervised baselines on the unseen Bamboogle benchmark with higher retrieval recall at comparable or fewer search steps, and a critic trained on Llama-3.1-8B raises GPT-4o-mini MedQA accuracy.
National Library of Medicine University of Alabama at Birmingham University of Virginia Yale University University of Illinois Urbana-Champaign
Qiao Jin（First author）Guangzhi Xiong（First author）Zhiyong Lu（Corresponding author）Aidong Zhang（Corresponding author）
RAG-Gym Re2Search++Process-level supervision Agentic Search Retrieval-Augmented Generation
arXiv Oct 2, 2026
MemLife builds entity-grounded first-person text memories read by a time-indexed agentic reader, beating the strongest training-free baseline by 4.6–12.0% on four long-horizon egocentric video benchmarks, with MemOpt adding 2.7–5.0% by training only the memory writer under the FIRM reward
The work introduces MemLife, a multimodal memory system that compacts egocentric video into time- and entity-anchored first-person text episodes retrieved by a time-indexed agentic reader, improving over the strongest training-free baseline by 4.6–12.0% across four long-horizon benchmarks without training or query-time video access, and further proposes MemOpt, which applies reinforcement learning only to the memory writer under the faithful, informative, and retrievable FIRM reward, adding 2.7–5.0% with gains that transfer across writer and reader backbones, memory systems, and out-of-domain video and question distributions.
The work introduces MemLife, a multimodal memory system that compacts egocentric video into time- and entity-anchored first-person text episodes retrieved by a time-indexed agentic reader, improving over the strongest training-free baseline by 4.6–12.0% across four long-horizon benchmarks without training or query-time video access, and further proposes MemOpt, which applies reinforcement learning only to the memory writer under the faithful, informative, and retrievable FIRM reward, adding 2.7–5.0% with gains that transfer across writer and reader backbones, memory systems, and out-of-domain video and question distributions.
University of Virginia University of Illinois Urbana-Champaign
Guangzhi Xiong（First author）
MemLife MemOpt FIRM Agentic Memory Video Question Answering

You’re all caught up

Reward the search, not just the answer: process supervision lifts retrieval recall and holds up out of domain

MemLife builds entity-grounded first-person text memories read by a time-indexed agentic reader, beating the strongest training-free baseline by 4.6–12.0% on four long-horizon egocentric video benchmarks, with MemOpt adding 2.7–5.0% by training only the memory writer under the FIRM reward