The paper investigates how providing execution traces to multimodal judges in agentic video‑generation systems can bias their verdicts. On a benchmark of 109 two‑event clips, traces that falsely report successful tool calls cause large‑language‑model judges to incorrectly accept 78–90 % of failures, while contradictory traces lead to 100 % rejection of correct clips. The effect persists even when judges are instructed to consider only the video frames, indicating that the vulnerability stems from the judges’ learned trust in tool logs rather than the visual content itself.
By Jian Xu
The paper introduces a method for co-evolving inspectable verifiers alongside self-improving agents. By synthesizing verifiers from clustered failures and selecting them based on agreement with a reference set, the authors demonstrate improved held‑out agreement on MBPP+ and outperform a bare LLM judge. They also show that removing anchor guards collapses the verifier into a vacuous grader, yet the collapsed verifier still trains skills effectively, indicating that downstream task scores cannot certify a self‑evolved verifier.
By Xing Zhang, Guanghui Wang, Yanwei Cui, Ziyuan Li, Wei Qiu, Bing Zhu, Peiyang He
ACE is a self‑correcting agentic canvas editor that operates on a hierarchical scene‑graph rather than flat document formats, enabling reliable multi‑slide presentation automation. It pairs a presentation‑specialized action space of 98 tools with CARE, a content‑aware router that reduces input tokens by about 89%, and a ground‑truth‑free instruction‑following judge that feeds natural‑language critiques back into the agent for self‑correction. In benchmarks, ACE outperforms a comparable agentic HTML pipeline on instruction following (4.23 vs. 3.81), runs 1.75× faster, costs 44% less, and is preferred by 58.7% of blind raters, with 81% favoring the self‑corrected output.
By JooYoung Jang, Taegyeong Lee, Jihyeon Park, Nojun Kwak
The paper investigates how a coding agent learns across a sequence of abstract reasoning tasks in the ARC-AGI-3 benchmark. The agent, operating on a frozen foundation model, writes and runs Python and shell scripts, retaining only its written artifacts as state. The authors develop a measurement protocol that traces each thought—any belief, rule, or plan committed to a file—through its lifecycle, revealing that most task-specific scripts are discarded at task boundaries and that the agent selectively forgets by rewriting knowledge into new scripts while accumulating unedited notes.
By Chen Wu, Josh Passenger, Yin Song
The paper introduces Randomized Block Curation (RBC), a method for selecting useful lessons from language‑model agents’ persistent textual memory. RBC evaluates the impact of grouped lessons through randomized experiments on practice tasks, retaining only those that demonstrably improve task success. Experiments across two memory systems, three models, and three benchmarks show RBC raises average success to 66.9, outperforming both full playbooks and agent‑chosen lessons.
arXiv:2610.02588v1 Announce Type: new
Abstract: Agents are increasingly given open-ended research tasks: discovering an empirical law from self-designed experiments, improving a heuristic whose optim...
By Chengyang Shi, Xianglin Ji, Jintao Huang, Jicheng Wang, Yifeng He, Jiachen Liu
Designer‑RSI presents a continual adaptation framework that lets a frozen frontier model operate professional design software while an external procedural memory learns natural‑language design skills from user traffic. Over five rounds on 1,406 real briefs and 1,869 graded trajectories, the memory grew from 76 to 139 skills, boosting execution success from 72.7% to 99.3% and improving win rates on four design benchmarks. The study shows that widening and deepening the memory, especially together, significantly outperforms a no‑skill baseline.
By Hongyang Du, Lan Yan, Christian Flores, Asim Kadav
arXiv:2607. 18064v1 Announce Type: cross Abstract: Coding agents can now be left alone to improve software against a score.
By Nursultan Askarbekuly, Mohamad Al Mdfaa, Ahmed Helaly, Gonzalo Ferrer, Manuel Mazzara
arXiv:2610.14216v1 Announce Type: cross
Abstract: Coding agents such as Claude Code also follow text files that developers write for them: a project file, skills the agent may choose to open, and rul...
By Samuele Giampieri
Professional graphic design is a long-horizon agentic task in which structured, editable artifacts emerge from many interdependent actions, yet outcomes admit no reliable programmatic oracle. We intro...
arXiv:2609.38119v1 Announce Type: new
Abstract: Long-form video understanding requires multimodal agents to iteratively gather evidence over many reasoning steps. However, most existing agentic metho...
By Jinfa Huang, Jianming Xu, Jingyang Lin, Zhengyuan Yang, Jiebo Luo
A language model's memory can be worse than having no memory at all. Give a model a memory that kept a wrong conclusion but dropped the work behind it, and it emits that stale value as a confident answer; give the same model an empty memory and it abstains.