The paper reports that applying reinforcement learning (RL) to the Kimi K2.7 Code model on 1,700 agentic coding tasks improves its performance on six external benchmarks. After a single epoch of GSPO training on a rank‑32 LoRA adapter, pass‑@1 scores increased across all benchmarks, with significant gains even on data released after training. The trained model also reduces agent steps and avoids common failure modes such as dropping requirements or breaking existing behavior.
By Sushant Mehta, Logan Ritchie, Edwin Chen
The paper investigates challenges in open‑source large‑language‑model (LLM) based multi‑agent systems (MAS). By analyzing 944 issues extracted from 21 projects, it finds that orchestration and execution problems are most common, with workflow, tool integration, and memory issues as primary causes. The predominant remedy identified is optimizing workflow, and the study offers empirically grounded implications for improving orchestration, tool integration, and memory mechanisms in LLM‑based MAS.
By Asad Ur Rehman, Syed Mohammad Kashif, Ruiyin Li, Peng Liang, Zengyang Li, Arif Ali Khan
The paper introduces SciUtopia, a closed‑loop large‑language‑model simulation framework that models the evolving academic research ecosystem, including research direction, collaboration, publication, peer review, funding, and researcher attrition. Running 61 simulation worlds, the system simulates over 40,000 researchers and 1.2 million LLM‑generated peer reviews, revealing that rejection‑driven resubmission increases reviewer burden, cautious exploration balances citation impact with career success, and resource inequality can arise without early‑funding advantage.
By Yiqiao Jin, Yiyang Wang, Lucheng Fu, Bing He, Siheng Xiong, Yijia Xiao, B. Aditya Prakash, Josiah Hester, Srijan Kumar, James Evans, Jindong Wang
The paper "Chaining Skills to Hijack LLM Agents" introduces APEX, a method that constructs adversarial skill chains to manipulate large language model agents into performing attacker-selected actions. By exploiting the sequential invocation of open-source skills, APEX creates a record of genuine task progress that falsely signals user approval, enabling downstream skills to carry out malicious actions. Experiments on SkillsBench and GPT‑5.4 show high success rates, and a prompting defense reduces but does not eliminate the threat.
By Tian Dong, Zixuan Ma, Haodong Zhao, Huaien Zhang, Shaofeng Li, Hao Chen
The paper surveys how decentralized agent economies combine independently designed protocols into a single task workflow, often leading to correct steps that still yield incorrect outcomes. It categorizes security and economic requirements into 17 property families across six stages, evaluates 12 systems and standards, and introduces the guarantee‑closure criterion to assess whether early guarantees persist to later stages. Empirical analysis of 840 executions and 11,648 checks reveals frequent failures between verification and settlement, highlighting gaps between recorded approvals and actual task conformance.
By Rui Sun, Xihan Xiong, Qin Wang, Fei Gao, Zelin Li, Zehua Cheng, Jiahao Sun, Zhipeng Wang
The paper introduces Token Communication (TokCom) as a native interface for Collaborative Embodied Artificial Intelligence (CEAI), where tokens act as compact semantic carriers and inference units for generative foundation models. It outlines a TokCom-assisted CEAI framework featuring a task‑adaptive communication protocol that includes a compact codebook, syntax rules, and contextual examples to guide token distillation and reconstruction over wireless channels. A case study on collaborative object transport shows that TokCom significantly reduces source payload bit consumption while maintaining task efficiency and robustness to noisy channels.
By Peng Yi, Ying-Chang Liang
This paper reviews blockchain-assisted intrusion detection and prevention systems for IoT and IIoT, noting that prior studies ignore the shift toward modern Endpoint Detection and Response (EDR) and Extended Detection and Response (XDR) architectures and lump blockchain’s roles together. It proposes a three‑axis taxonomy—detection‑system class, blockchain functional role, and response‑automation maturity—to classify recent research (2019‑2026). The authors identify gaps such as the lack of per‑endpoint blockchain‑anchored response loops due to latency, deployment, and community mismatches, and highlight cross‑cutting challenges like consensus latency, post‑quantum cryptography, smart‑contract vulnerabilities, and LLM‑based detection risks, concluding with a research agenda for hybrid on‑chain/off‑chain orchestration.
By Yahya Shahsavari, Sara Rouhani, Kaiwen Zhang
arXiv:2604.01375v3 Announce Type: replace
Abstract: Rubrics distill notions of expert quality and measure agent performance. However, the quality of rubrics themselves have not been systematically me...
By Ankit Aich, Zhengyang Qi, Charles Dickens, Derek Pham, Esha Sharma, Josh Viktorov, Amanda Dsouza, Armin Parchami, Frederic Sala, Paroma Varma
DuoMind is a distributed hierarchical framework that enables multi-robot coordination by combining vision-language-action models for low‑level execution with vision-language models for high‑level reasoning and inter‑robot communication. Each robot’s orchestrator processes task instructions, local observations, and messages from peers to generate precise action commands and semantic messages for other robots. The authors introduce RoboPoly, a benchmark of long‑horizon manipulation tasks, and show through experiments that DuoMind improves multi‑robot task performance, with ablation studies highlighting the roles of hierarchical orchestration and semantic communication.
By Hanchu Zhou, Dechen Gao, Hang Wang, Brendan Lynch, Boqi Zhao, Qiyao Ma, Raman Goyal, Junshan Zhang
KaliBench is a fine‑grained benchmark that evaluates large language models’ ability to translate natural‑language queries into executable command‑line instructions for Kali Linux. It contains 8,504 query–command pairs covering 1,642 tools across 23 capability dimensions and 5 security phases, and uses a deterministic, alias‑aware evaluation pipeline with multi‑stage verification (LLM validation, sandboxed execution, and human refinement). The benchmark also supports runtime‑free verifiable rewards, enabling training improvements that bring smaller models close to the performance of a 685B MoE model.
By Pengfei Li, Naufal Suryanto, Sicheng Zhang, Muzammal Naseer
The paper introduces MSARL, a Multi‑Small‑Agent Reinforcement Learning framework that separates reasoning from tool use. A single Reasoning Agent plans and decomposes tasks, while multiple specialized Tool Agents handle specific external tools, each trained with imitation learning and reinforcement learning using role‑specific rewards. Experiments on mathematical problem solving with code execution show that MSARL improves reasoning stability and final‑answer accuracy compared to single‑agent baselines, and the architecture generalizes to other tool‑use tasks.
By Dayu Wang, Yutong Liu, Jiaye Yang, Weikang Li, Jiahui Liang, Yang Li
The paper introduces a framework for auditing grounding claims in AI systems, evaluating how tokens like "cat" are linked to real-world concepts. It reports measurements, evidence, and verdicts based on explicit acceptance criteria, assessing accuracy, robustness, and composition while documenting how mechanisms were acquired and retained. A toy gridworld example shows an agent correctly interpreting individual symbols but failing a withheld combination, highlighting a deviation from the declared composition rule and demonstrating the audit’s ability to pinpoint such issues.
By Daniel Quigley, Eric Maynard
AstroAgentBench is a seven-family benchmark designed to evaluate agentic planning for space mission tasks such as scheduling, observation planning, constellation design, and relay support. The benchmark requires agents to submit planning artifacts that are verified for schema, timing, geometry, resources, and mission value, with results reported as validity and normalized scores compared to task-specific solver references. Across five LLM agent systems and 35 held-out cases, the strongest systems approach or exceed solver-reference scores on several families, while weaker systems often fail to produce high-value valid plans, and even strong systems lose quality on geometric, product-level, or design-heavy tasks. Failure analyses identify task-contract misformulation and weak solution construction as key issues, and successful runs demonstrate that agent calibration against verifier feedback and adaptive search can improve performance; ablation studies show that procedure injection and memory accumulation help when they provide the missing formulation, calibration, or search support.
By Weiyi Wang, Xinchi Chen, Jingjing Gong, Xuanjing Huang, Xipeng Qiu
PhGPO: Pheromone-Guided Policy Optimization for Long-Horizon Tool Planning proposes a new method for improving multi-step tool planning in large language model agents. The approach draws inspiration from ant colony optimization, learning a trajectory-based transition pattern—referred to as pheromone—from historically successful tool-use paths. This learned pheromone is then used to guide policy optimization, steering the agent toward reusable tool transitions and enhancing long-horizon planning performance.
By Yu Li, Guangfeng Cai, Shengtian Yang, Han Luo, Shuo Han, Xu He, Dong Li, Lei Feng
The paper introduces IMAVB, a 500‑clip benchmark that tests whether omnimodal large language models can detect when a textual premise contradicts their visual or audio input. Experiments on eight open‑source models and Gemini 3.1 Pro reveal a Representation‑Action Gap: internal states encode mismatches, yet the models rarely reject false premises, exhibiting under‑rejection or over‑rejection. A probe‑guided logit adjustment improves rejection behavior, suggesting the main bottleneck is in translating perception to action rather than in perception itself.
By Trung Nguyen Quang, Yiming Gao, Fanyi Pu, Kaichen Zhang, Shuo Sun, Ziwei Liu
The study investigates how different observation fidelities affect the problem‑solving performance of embodied large language model (LLM) agents. Using the Lockbox mechanical puzzle, agents were tested with raw RGB, RGB‑D, and perfect ground‑truth symbolic observations in both physical and simulated environments. Surprisingly, agents performed best with raw RGB input and worst with perfect observations; introducing moderate perceptual noise in simulation further improved success rates, suggesting that some errors can mitigate repetitive action loops.
By Oussama Zenkri, Oliver Brock
The paper presents a reinforcement learning framework for training autonomous large language model agents to act as sellers in multi-product bargaining markets. It models the problem as a Partially Observable Markov Decision Process and introduces a structured message protocol to map natural language into a regulated decision space. Using Reinforcement Learning from Verifiable Rewards, the trained agents match limited inventory to buyers more effectively, achieving seller surplus extraction and allocation quality comparable to or better than trillion‑parameter models, and generalize to unseen market conditions.
By Shuze Daniel Liu, Claire Chen, Jiuqi Wang, David Simchi-Levi, Thorsten Joachims
VERA is a centralized‑training, decentralized‑execution framework for constrained multi‑agent control that separates feasibility estimation from credit assignment. Each actor predicts a five‑dimensional verifiable feasibility representation (VFR), which is supervised by exact action‑conditioned margins during training and ranked by a counterfactual group‑relative advantage (CGRA). In a dynamic space‑air‑ground integrated network, VERA achieves over 55% success with minimal coverage violation, outperforming baselines by up to 8.74 percentage points and reducing violations by more than 50%.
By Bo Yin, Dongbo Li, Hongkai Chen, Jie Liu, Guoliang Xing
The paper introduces VirusCascade, a black‑box attack that exploits the collaborative reflection mechanism in LLM‑powered recommender agents. By leveraging reflective persistence and cross‑agent propagation, the attack promotes a target item through semantic rationalisation and structural positioning, achieving state‑of‑the‑art targeted exposure on four real‑world datasets. The study demonstrates that existing static‑pipeline attacks cannot exploit this recurrent, multi‑agent amplification pathway.
By Yurong Hao, Wen Zhou, Guowei Guan, Tiantong Wu, Fuyao Zhang, Wei Yang Bryan Lim
Diffusion-2BC is a hybrid training method that combines a diffusion denoising objective with an auxiliary deterministic behavior‑cloning loss over a shared visual encoder. The auxiliary loss is used only during training, while inference remains diffusion‑based. Experiments on the Claw environment and CARLA navigation show that Diffusion‑2BC reduces mean mask‑distance error by about 10% compared to a diffusion baseline and by 85% compared to standard deterministic behavior cloning, and it enables the agent to travel farther and exhibit multimodal route choices.
By Bruno Maciel Machado, Eric Aislan Antonelo