We study what happens when a single general-purpose large language model acts as the sole researcher on a long-horizon neural architecture design problem. The agent receives a scientific question, an initial hypothesis and motivation, a compute budget, and research affordances (source and experiment management, experiment tracking, literature access, and persistent memory), then autonomously proposes, implements, evaluates, and records experiments over an extended period.
On-policy distillation (OPD) provides teacher supervision on states visited by the student, reducing the distribution gap between training and inference. However, in multi-turn agentic tasks, student deviations may accumulate over time, gradually moving the trajectory away from states where teacher guidance remains effective.
Large language models (LLMs) are increasingly extended into deep search agents that solve complex questions through multi-step interaction with external search and browsing tools. However, existing agents often incur substantial computational and interaction costs, generating lengthy trajectories that contain redundant queries, inefficient exploration, and irrelevant observations.
As memories accumulate across tasks and sessions, the performance of long-term LLM agents depends jointly on query-specific retrieval and continual memory refinement. However, existing methods typically optimize either memory access, through iterative query refinement or adaptive retrieval policies, or memory evolution such as structural update.
Multi-agent large language model (LLM) systems can expose protected state through internal messages, tool arguments, logs, and persistent memory even when their public outputs appear innocuous. Existing privacy prompts, redaction methods, and source-level access controls restrict surface content or data access, but do not specify what a legitimately informed agent should disclose or how that disclosure may be reused downstream.
Verifier-guided reinforcement learning has become a powerful paradigm for improving LLM reasoning. In multi-turn settings, models receive a verifier score after each turn and iteratively refine their outputs.
arXiv:2607. 29617v1 Announce Type: cross Abstract: Imitation learning (IL)---training an agent to replicate expert behavior from demonstrations---underpins applications from robotics to language model training.
By Luca Viano, Antoine Moulin, Audrey Huang, Volkan Cevher, Philip Amortila, Dylan J. Foster
arXiv:2607. 29241v1 Announce Type: cross Abstract: Optimizing modern recommender models still depends heavily on engineers manually iterating over architectural, objective, and training-strategy changes.
By Haoran Ling, Yuecheng Li, Zeyu Song, Jing Yao, Shuwen Kang, Chi Lu, Wenjin Wu, Peng Jiang
arXiv:2607. 29419v1 Announce Type: cross Abstract: In reinforcement learning, exploration with sparse and delayed rewards presents a significant challenge due to the limited feedback available for guiding the learning process.
By Bumgeun Park, Donghwan Lee
arXiv:2607. 29422v1 Announce Type: cross Abstract: Automated vulnerability repair aims to reduce the time and effort required to patch security flaws from a vulnerability triage report.
By Michael Fu, Qiyue Mei, Patanamon Thongtanunam, Kla Tantithamthavorn
arXiv:2407. 04521v3 Announce Type: replace-cross Abstract: This paper studies the continuous-time q-learning in mean-field jump-diffusion models in a setting where the environment simulator does not provide direct access to the population distribution.
By Xiaoli Wei, Xiang Yu, Fengyi Yuan
arXiv:2607. 28881v1 Announce Type: new Abstract: As more responsibility is placed upon AI systems, it becomes increasingly important to guarantee that these systems are aligned with humanity.
By Winter Cross
arXiv:2607. 29181v1 Announce Type: cross Abstract: Agentic assistants capable of proactive, personalized interactions require structured models of user intent and workflow.
By Andy J. Phu, James Mooney, Karin de Langis, Khanh Chi Le, Dongyeop Kang
arXiv:2603. 20075v2 Announce Type: replace-cross Abstract: Compilers are critical to modern computing, yet fixing compiler bugs is difficult.
By Yingwei Zheng, Cong Li, Shaohua Li, Yuqun Zhang, Zhendong Su
arXiv:2607. 29218v1 Announce Type: new Abstract: With the prosperity of the large language models (LLMs), it has become an interesting topic: how do LLM-based agents work in Minecraft?
By Jianxin Gao, Beini Hu, Runze Li, Wanli Peng, Ruohan Lei, Jinyuan Zhang, Linna Deng, Tianyi Yu, Zining Wang
arXiv:2605. 07699v2 Announce Type: replace-cross Abstract: LLM-based agents are increasingly deployed for routine but consequential tasks in real-world domains, where their behavior is governed by inherently ambiguous domain policies that admit multiple valid interpretations.
By Hsuvas Borkakoty, Sebastian Pohl, Cheng Wang, Bei Chen, Yufang Hou
arXiv:2607. 28631v1 Announce Type: new Abstract: AI Scientist systems capable of autonomous research have the potential to significantly accelerate scientific discovery.
By Vaibhava Lakshmi Ravideshik, Mayank Kejriwal
arXiv:2607. 28678v1 Announce Type: new Abstract: Multimodal agents operating in long-horizon environments must build and continually update multimedia memories to support entity-consistent, temporally grounded reasoning.
By Xinkui Zhao, Enbo Chen, Yifan Zhang, Chang Liu, Guanjie Cheng, Naibo Wang, Yueshen Xu
arXiv:2607. 28990v1 Announce Type: new Abstract: Large language model agents have shown promising capabilities in data-driven scientific discovery tasks, where an agent interacts with an execution environment and produces a statistical claim.
By Yucheng Xu, Keyi Zhang, Yuyang Yu, Min Zhang, Shiyuan Meng, Pei Chu, Zhongying Tu
arXiv:2607. 29516v1 Announce Type: cross Abstract: AI coding agents are generating code at volumes that exceed the capacity of traditional peer review.
By Chandra Maddila, Mashrur Rashik, Euna Mehnaz Khan, Smriti Jha, James Saindon, Nachi Nagappan, Peter C. Rigby