An agent that uses tools typically responds to what the user explicitly asks, yet completing the task may require information the user never requested. Work on proactive agents mainly studies whether...
Users of language-based assistive agents often make ambiguous requests. In response, an assistant can either directly act on its interpretation of the request --- risking misalignment with the user --...
The paper investigates how different retrieval designs affect conversational agents. Three approaches are compared: skill-based retrieval in the main agent, a non‑persistent retrieval subagent that generates findings, and a persistent retrieval subagent that keeps history and returns evidence to the main agent. The persistent design reduces token usage and API cost by 43.3% and 45.0% respectively, cuts response time by 9.1%, and allows the main agent to request corrections from the subagent.
By Karthik Ravula, Sathwik Kowtha, Rishabh Verma, Kala Mehak Jain, Gunjit Mittal
arXiv:2609.37588v1 Announce Type: new
Abstract: Users of language-based assistive agents often make ambiguous requests. In response, an assistant can either directly act on its interpretation of the...
By T. Duy Nguyen-Hien, Yee Whye Teh, Wee Sun Lee, Tan Zhi-Xuan
arXiv:2608. 07531v1 Announce Type: cross Abstract: Search-augmented language agents should retrieve external information only when necessary and ground their answers in retrieved evidence.
By Cheng Ruoxi, Ma Haoxuan, Zhang Hongyi, Zhang Junming, Duan Ranjie, Xia Qiaolin, Wang Hao, Lu Yu, Shi Haibo, Ma Xingjun
arXiv:2610.35083v1 Announce Type: new
Abstract: AI assistants are increasingly asked to reason over a user's own web traces: group chats, community threads, activity logs and revision histories. On C...
By Jinhu Qi, Minda Hu, Wentao Zhang, Zixing Song, Irwin King
The paper introduces the concept of an agent’s "taste"—its ability to make effective long‑horizon decisions—and presents Taste‑Bench, a new benchmark that automatically generates decision‑fork questions from agent trajectories. Taste‑Bench evaluates models on choosing the best path without seeing future outcomes, revealing that top models answer only about 60% of questions correctly and that later‑appearing evidence makes forks harder. The authors also demonstrate that training a student model to mimic a teacher’s judgment improves decision quality and overall success on held‑out software engineering tasks.
By Wenbo Pan, Zhichao Liu, Shujie Liu, Jingying Zeng, Chin-Yew Lin, Xianfeng Tang, Yan Lu, Qi He, Xiaohua Jia
arXiv:2608.01822v2 Announce Type: replace
Abstract: Training LLM-based search agents requires high-quality search data: tasks that demand genuine multi-hop retrieval and trajectories that use search...
By Wentao Tan, Qiong Cao, Jiaqi Wang, Nan Duan
The paper introduces FoG, a framework that trains knowledge graph (KG) agents to retrieve the evidence behind their answers. FoG combines evidence‑grounded synthesis, a KG interaction agent, and an evidence‑coverage process reward (ECPR) to ensure that every answer is supported by retrieved triples. Experiments on three KGQA benchmarks show that 3B–4B open‑source agents trained with FoG outperform larger baselines in both answer accuracy and evidence coverage, even with zero‑shot cross‑KG transfer.
By Zongyuan Sun, Shuai Zhao, Yingjun Du, Ke Shu, Ziyue Zhu, Haoran Luo
The paper examines three retrieval designs for conversational agents: skill-based retrieval in the main agent, a non-persistent retrieval subagent that generates findings, and a persistent retrieval subagent that keeps retrieval history and returns evidence directly to the main agent. The persistent design reduces token usage by 43.3% and API cost by 45.0% compared to skill-based retrieval, while cutting average response time by 9.1%. Compared to non-persistent retrieval, it speeds up responses by 15.6 seconds per question but uses 65.1% more tokens and costs 29.1% more.
arXiv:2610.20959v2 Announce Type: replace-cross
Abstract: Answering questions over heterogeneous data means combining evidence from files, databases, documents and the web. Data agents automate this,...
By Zhivar Sourati, Marianne Menglin Liu, Zheng Wang, Rongguang Wang, Sai Ashish Somayajula, Tao Sheng, Kyu J. Han, Sujith Ravi, Dan Roth
The study examines how three context‑engineering techniques—semantic layers, orchestration, and compaction—affect data agents that answer questions over heterogeneous data. Using seven large language models on 1,514 long‑horizon tasks, the authors identify three agent behavior groups (investigators, premature answerers, early stoppers) and show that each technique benefits only the group it aligns with, while sometimes harming others. Combining the techniques yields smaller net gains than the sum of individual effects, leading to guidance on selecting techniques based on agent behavior, task demands, and accuracy‑cost trade‑offs.
By Zhivar Sourati, Marianne Menglin Liu, Zheng Wang, Rongguang Wang, Sai Ashish Somayajula, Tao Sheng, Kyu J. Han, Sujith Ravi, Dan Roth