The paper introduces TacEx, a tactile‑curiosity framework that guides reinforcement learning agents to explore contact dynamics by focusing epistemic uncertainty on the tactile channel. By anchoring curiosity to touch, robots learn to manipulate and grasp objects without task rewards or demonstrations, generating an interaction‑dense dataset that supports offline pick‑and‑place policy learning. TacEx also enhances vision‑language‑action models through post‑training, improving downstream performance while remaining sample‑efficient.
By Klemens Iten, Alexander Proshkin, Bhavya Sukhija, Stelian Coros, Andreas Krause, Pieter Abbeel, Carmelo Sferrazza
arXiv:2609.38201v1 Announce Type: new
Abstract: Long-running tools can dominate coding-agent latency: compilers, test suites, and repository commands take seconds to minutes while the agent idles. Th...
By Jiangnan Yu, Ceyu Xu, Mengming Li, Shiyu Huang, Yiran Xia, Jian Weng, Hui Xue, Haohui Mai, Yuan Xie
The paper introduces the QuantCode Model, a specialized language model for generating executable algorithmic trading code. It combines continued pretraining on trading framework code with supervised fine‑tuning on validated request‑to‑code pairs, achieving significant gains on the QuantCode‑Bench benchmark and improving both single‑turn and multi‑turn agentic success rates. The study also highlights the importance of retaining parsing and tool‑calling capabilities during domain specialization.
By Alexey Chernysh, Orkhan Ekhtibarov, Dmitry Zmitrovich
MetaSteer is a new method for steering large language models that learns nonlinear, context-dependent interventions applied to attention projection matrices. Unlike traditional linear, context-independent techniques, MetaSteer adapts its effects based on the input, requiring no linear concept-geometry assumption. Trained once on a pooled preference corpus, it transfers zero‑shot to unseen concepts and out‑of‑distribution contexts, matching or surpassing strong task‑specific baselines on multiple benchmarks and model families.
By Mehdi Jafari, Hao Xue, Flora Salim
The paper introduces AutoMIP, a reusable agent skill designed to manage long‑horizon autoresearch for mixed‑integer linear and nonlinear programming problems. AutoMIP combines idea pooling with an algorithm tree search to maintain a persistent pool of complementary candidate ideas and organize experiments into a structured tree, allowing the agent to preserve unexplored hypotheses, refine promising algorithms, and switch methodological directions based on historical states. Experiments on MILP and MINLP benchmark cohorts show that AutoMIP achieves the highest final success rates, discovering new best solutions for 31 of 60 MIPLib instances and 52 of 60 MINLPLib instances, and ablation studies confirm the complementary contributions of idea pooling and algorithm tree search.
By Yuwei Gu, Yaoxin Wu, Tong Guo, Wen Song, Zhiguang Cao
Cogentic is a multi‑agent system designed to automate proof discovery for open research problems. It uses an iterative prove‑verify loop where an orchestrator assigns independent provers to different proof directions, verifies their outputs with specialized components, and records confirmed intermediate results in a persistent ledger for future rounds. Built on Gemini, Cogentic has produced novel results on five open problems in online learning, auction theory, and mechanism design, each verified by domain experts and detailed in companion papers.
By Yang Cai, Vineet Gupta, Yanchen Jiang, Christopher Liaw, Aranyak Mehta, Grigoris Velegkas, Di Wang
arXiv:2307.15494v3 Announce Type: replace-cross
Abstract: Hindsight Experience Replay (HER) enhances sample efficiency in goal-conditioned reinforcement learning (RL) by relabelling failed trajectori...
By Kevin Yandoka Denamgana\"i, Daniel Hernandez, Ozan Vardal, Sondess Missaoui, James Alfred Walker
MADBench is a benchmark that evaluates the security of Multi-Agent Debate (MAD) systems, which allow large language models to exchange and critique answers. The study categorizes attacks into a layered taxonomy aligned with the MAD workflow and tests six attack families across 356 source tasks and 3,958 test cases. Results indicate that while MAD can reduce answer accuracy attacks compared to single-agent baselines, it may amplify unauthorized reads or writes, and even with collusion among agents, the final answer changes from correct to wrong only 28.30% of the time.
By Yuwan Liu, Jiaming Zhang, Yue Huang, Sisi Duan
RefCon is a method for extracting higher‑quality memories in context‑evolving agents by combining sequential self‑refinement with parallel self‑contrast, without relying on gold labels. It shows strong, consistent gains on benchmarks such as AppWorld, BFCL‑V3, ACE, ReMe, and ReasoningBank, outperforming existing baselines across different model scales and even surpassing ground‑truth baselines in software engineering tasks. The approach also maintains favorable efficiency and continues to improve as more trajectories are used, unlike diversity‑only scaling which saturates earlier.
By Ubaidillah Ariq Prathama, Bo Liu, Yeo Boon Hong, Yu-Xuan Huang, Yangkai Ding, Tao Yu
The paper introduces EvolvingNav, a system that builds a time‑indexed belief about moving targets in dynamic environments by combining timestamped 3D object histories with a persistence‑relocation model. It uses an event‑driven filter to update beliefs over time, incorporates RGB‑D evidence, and applies a zero‑shot vision‑language controller for action selection. The authors also present EvoWorld‑Bench, a large benchmark of human‑activity‑based scenes, and demonstrate that EvolvingNav outperforms baselines in both simulation and real‑robot experiments, especially when temporal patterns are learnable.
By Mingjian Gao, Zhaocheng Li, Haoyang Huang, Wenqiao Zhang, Yingjie Niu, Hao Zhou, Chao Li, Juncheng Li, Siliang Tang, Yueting Zhuang
AVERT-VLN introduces a closed‑loop framework for vision‑and‑language navigation that incorporates an abstention‑aware Monitor to detect instruction‑execution inconsistencies. The Monitor is trained on a new LOSTNAV dataset of 20K counterfactual risk trajectories and fine‑tuned on 40K normal trajectories to recognize semantic deviations. During deployment, the Monitor can suspend autonomous navigation and request human guidance, while offline preference learning uses deviation‑associated failures to improve the policy, achieving 76.2% and 66.3% success on R2R‑CE and RxR‑CE unseen splits.
By Minrui Liu, Jingke Wang, Yuehao Huang, Hao Su, Jiajun Lv, Yukai Ma, Yong Liu
The paper examines how AI companion agents describe themselves and how users can verify these claims. It argues that users can only confirm self-descriptions when evidence is directly observable in the agent’s behavior or in the user’s own experience, not when it resides within the agent’s internal mechanisms. The authors present an audit procedure that compares an agent’s self-descriptions with user judgments and implementation records, and apply it to a deployed companion named Lita, finding that users endorsed stylistic claims but were skeptical of relational claims, and that some memory components never executed as claimed.
By Seiya Ikeda, Shin-nosuke Ishikawa
The paper investigates a critical flaw in AI coding-agent systems such as Claude Code, Codex CLI, and Cursor, where the action approved by a human is not the same as the action executed by the harness. It introduces the concept of Approval Laundering, categorizing six systematic failure modes—Scope, Argument, Temporal, Tool, Delegation, and Semantic laundering—and demonstrates these failures through controlled experiments. The authors propose an Approval Token mechanism that mitigates some laundering types but leaves others unaffected, highlighting the limitations of current enforcement strategies.
By Yang Wang
The paper evaluates an agentic harness that combines a 27B language model, financial calculations, narrative templates, and validation checks to produce corporate and investment banking presentation decks. Using a panel of five judges, the system consistently scores higher than a baseline model that generates directly from a short prompt, with scores ranging from 20.4 to 33.6 out of 95. However, judge variability and changes in grading criteria make it challenging to discern small improvements in deck quality.
By Ludovic Gibert, Matis Despujols, Andre-Louis Rochet
The paper evaluates hard‑gate candidacy for validators in a deployed generative‑agent system by measuring how well each validator’s firing separates successful from failed builds. Across 13 validators and thousands of builds, only a few checks show statistically significant separation, while many fail to distinguish or never fire. The study highlights that skipped checks are recorded as passes, limiting detectable failure rates and underscoring the need for clearer evaluation records.
By Xin Xu
The authors report that a large‑scale experiment using a language‑model coding agent over five weeks produced new lower bounds for DNA‑barcode‑style codes. By restricting searches to codes with a prescribed symmetry, the agent improved the best known code of length 6 and minimum edit distance 3 from 114 to 120 words, and similarly raised lower bounds for lengths 6–9 and distances 3–6. The study also documents failures and the limitations of the verification protocol, noting that intermediate results were never rechecked and could lead to erroneous conclusions.
By Abraham Yeung
The paper presents a controlled comparison of six retrieval-augmented generation (RAG) strategies for scientific question answering on a large arXiv corpus. All pipelines use the same LLM generator and evaluation protocol, differing only in retrieval design—ranging from classic dense retrieval to late‑interaction methods like ColBERTv2. The authors also release a synthetic question dataset and code to enable reproducible, large‑scale evaluation of RAG trade‑offs.
By Bhagyesh Rathi, Eshan Chawla, William B. Andreopoulos
arXiv:2609.39378v1 Announce Type: new
Abstract: Real-world embodied tasks, from everyday activities to professional procedures, require agents to act under physical constraints while tracking evolvin...
By Shulin Tian, Junsu Kim, Shuai Liu, Hao Li, Yujiao Shen, Sihan Li, Zhe Yang, Yeongon Kim, Feiyu Li, Jialin Wu, Yichi Zhang, Wenhui Wang, Runmao Yao, Yuhao Dong, Zhaoxi Chen, Fangzhou Hong, Antonino Furnari, Jingkang Yang, Hongyuan Zhu, Ziwei Liu
OpenCollab is a multi‑agent coding framework that unifies organization design, enforces experimental control on a shared runtime, and tracks execution via fine‑grained event streams. It introduces the metric Adherence to measure whether the declared organization is actually realized, showing that small configuration changes can shift adherence from 47.2% to 97.2%. Experiments demonstrate that a two‑coder workflow built on OpenCollab achieves new state‑of‑the‑art performance against mainstream harnesses while using the fewest tokens, and that a well‑designed organization can outperform strong existing harnesses.
By Chun-Wah Hsu, Kai Gong, Yu Wu, Xianhe Chen, Mengyang Liu, Jie Li, Hanyu Li, Zhixuan Liu, Naisheng Tang, Jiaying Chi, Ziheng Fan, Xuning He, Xiaokang Yang, Xue Jiang, Yihong Dong
The paper introduces AgenticShadow, a new dataset of 17,138 image‑mask‑target triplets created through an offline agentic workflow that combines physics‑motivated generation, failure detection, feedback‑driven retry, candidate selection, and deterministic correction. This approach addresses the long‑standing lack of diverse paired shadow‑free training data by leveraging existing shadow detection datasets and producing realistic shadow‑free targets. Models trained on AgenticShadow show significant improvements, reducing color distribution differences by 50.5% and cross‑domain LAB RMSE by 19.7‑37.5% compared to prior work.
By Shilin Hu, Jingyi Xu, Dimitris Samaras, Hieu Le