The paper introduces Humanize MLE Agents (HMA), a workflow that alternates two agents across sessions on a shared workspace, limiting each session to at most k experiments and resetting context between sessions. This approach, modeled as a context‑reset semi‑Markov decision process, outperforms single‑agent baselines, achieving a 78.2% any‑medal rate on 75 Kaggle MLE‑bench tasks—8.0 points higher than the best open‑source baseline and using only a quarter of its time budget. Across six heterogeneous agent pairs, HMA improves the any‑medal rate by an average of 6.4 points, and a supplementary kernel‑optimization case reaches 878 cycles, within 1.6% of the community leaderboard leader.
By Changye Li, Zijian Zhang, Muheng Li, Yiduo Pang, Tingjun Wu, Menghan Li, Yitong Liu, Hongkai Guo, Mengchi Chen, Chenyu He, Kaisheng Ma, Wei Xu, Song Han, Sihao Liu, Ligeng Zhu
The paper introduces a multi‑agent large language model framework that automatically tunes key components of specialized neural network models without altering their overall architecture. Three agents—analyst, coder, and recorder—operate within a sandboxed environment to generate code for specified components, enabling efficient optimization across diverse tasks. Experiments on spiking neural networks for classification and Physics‑Informed Neural Networks for solving high‑order differential equations demonstrate significant performance gains, such as 13‑point accuracy improvements on MNIST and substantial reductions in error metrics for nonlinear Schrödinger and Kuramoto–Sivashinsky equations.
By Ivan Moloshnikov, Dmitry Kunitsyn, Alexander Sboev, Maksim Skorokhodov, Roman Rybka, Yuanchao Liu
X-AgentRed is a self‑evolving red‑teaming framework designed to uncover harmful tool‑call paths in personal agents that interact with multiple web services. It builds a directed graph of tool dependencies by analyzing each tool’s effects, potential harms, and execution preconditions, then samples realistic scenarios from this graph. Using execution feedback, the framework diversifies harmful paths through tool substitution and refines blocked steps, achieving a 1.77× higher harmful path rate than STAC and 2.99× greater tool coverage across four frontier agentic models.
By Yeonsung Jung, Zhiyuan Wang, Zifan Wang, V\'itor Albiero, Benjamin Newman, Edoardo Debenedetti, Ivan Evtimov, Ninareh Mehrabi
RATED (Reliability‑Aware Tool Retrieval via Execution Dynamics) is a retrieval framework that blends robust semantic matching with dynamic reliability adaptation for AI agents using external tools. It uses a frozen dual‑encoder with contrastive fine‑tuning for semantic relevance and a lightweight contextual bandit module that re‑ranks tool candidates based on real‑time execution feedback, allowing rapid adaptation to runtime failures without costly retraining. The EcoAgent benchmark, comprising 5,172 real‑world tools and over 210,000 query‑tool pairs, shows that RATED improves mean reciprocal rank by 52.2% over the ToolLLM API retriever and cuts fault selection from 86.2% to 2.9% during severe API outages, with negligible latency overhead.
By Weiliang Chen, Song Liao
The paper presents iRAG, a human‑in‑the‑loop decision‑support system that blends Retrieval‑Augmented Generation with Hybrid Decision‑Making. Unlike traditional RAG models that rely on static knowledge bases, iRAG starts with an empty KB and incrementally builds it from human decisions, storing past decisions as retrievable precedents. The system progresses from silent observation to evidence‑grounded contestation and ultimately to autonomous handling with abstention, achieving low error rates even under knowledge drift and demonstrating a stability–plasticity trade‑off.
By Federico Mazzoni, Andrea Failla
The paper presents WZDrive, a multimodal ego‑trajectory predictor for autonomous vehicles operating in urban road construction zones. It introduces WZDrive‑Data, a dataset of 1,183 work‑zone driving segments collected from 33 drivers in a human‑in‑the‑loop simulator that records driver control, gaze, vehicle trajectories, and scene information. WZDrive achieves a mean minADE of 1.09 m and minFDE of 2.12 m, outperforming state‑of‑the‑art baselines by 7.18 % and 13.18 % respectively.
By Shuo Zhang, Aman Sunesh, Semiha Ergan, Kaan Ozbay
Meta-Reflective Evolution Effectively Improves Production AI Systems describes CoCoEvolve, an evolutionary optimizer that uses meta-reflection to retrieve evidence from the entire evolutionary history and probe live systems before proposing mutations. The approach is applied to data agents, LLM-powered data pipelines, and data engineering workflows, achieving state‑of‑the‑art results on DABStep, multiple LLM pipeline workloads, and data-eng-bench. Ablation studies confirm that each component of CoCoEvolve contributes to its performance gains.
By Seung Won Wilson Yoo, Chang Zeng, Aditi Somayajula, Sahithi Ankireddy, Rohan Kosalge, Avishree Khare, John Mitchell, Aditya G. Parameswaran, Rajhans Samdani, Anupam Datta
Agent4Spec is an LLM-based autonomous multi‑agent system designed to improve API specifications through continuous validation and market analysis. It includes a validation pipeline that detects and self‑corrects OpenAPI syntax errors, and a market analysis pipeline that performs competitive research, identifies functional gaps, and automatically generates missing API‑spec functionalities. In evaluations on curated datasets, the best LLM achieved 100% syntax error detection, over 95% success in implementing corrections, and more than 50% adoption of suggested endpoint improvements.
By Yongjian Tang, Costel Moraru, Doruk Tuncel, Hanya Elhashemy, Steffen Klepke, Thomas Runkler
PRIME introduces an index-side optimization framework for structured document retrieval, comprising a Semantic Restructuring track that converts long documents into dense, structured sections and a Metadata Synthesis & Hierarchical Retrieval track that extracts semantic anchors and heading structures for efficient coarse-to-fine retrieval. The approach reduces embedding computation by 45.8× and index vector count by 22.4× while maintaining retrieval quality comparable to structure-aware chunking. In full deployment, Semantic Restructuring improves Answer‑F1 by 13.7%, and the study identifies qualitative regimes that influence when index-side optimization is beneficial or detrimental, noting that strong rerankers neutralize these effects.
By Cheuk Hei Chung, Li Liu
ProgramGym introduces a scalable framework for training long-horizon program reproduction agents, comprising Program-Env and Program-Agent components. Program-Env builds verifiable training environments from repositories via a feedback loop that repairs sandbox builds and improves black-box tests, yielding 1,648 environments across C, C++, Go, and Rust. Program-Agent employs a role-isolated workflow of analysis, implementation, and review with revision management, boosting performance on ProgramBench and raising Qwen3.8-27B’s accuracy from 34.87% to 42.84% and surpassing flagship models with fine-tuning.
By Chengyi Yang, Renxu Bao, Jue Hou, Zongpei Teng, Ziheng Wang, Yujie Lin, Chengsong Huang, Junnan Zhu, Delai Qiu, Yining Wang, Shengping Liu, Fei Long, Jinsong Su
The paper proposes adding a social component—marginal cost—to autonomous vehicle (AV) routing rewards to account for the impact of AVs on human-driven vehicles (HDVs). Using multi-agent reinforcement learning on real-world networks, the authors show that this social reward reduces HDV travel time increases and improves overall system performance compared to purely selfish AV routing. A scalable approach limits counterfactual simulations to the most influential AVs, maintaining performance while cutting computational cost.
By Anastasia Psarou, Mikolaj Rams, Blazej Torbus, Rafal Kucharski
SafeDiSTL is a constrained multi‑agent reinforcement learning framework that decomposes joint constraints using Signal Temporal Logic robustness to assign credit to individual agents. By measuring episode‑level violations through running‑maximum cost increments and computing action sensitivities, the method distributes each cost increment among agents at the same decision time, accounting for delayed effects. Experiments on cooperative payload transport, connected navigation, and networked traffic control show that SafeDiSTL improves safe‑success rates and throughput compared to constrained baselines while preserving unconstrained policy performance.
By Hongrui Zheng, Aditya Singh, Rahul Mangharam
AutoControl Arena is an automated framework for evaluating frontier AI risk that separates deterministic state from generative dynamics, reducing logic hallucination while preserving flexibility. The system uses a three‑agent architecture and achieves over 98% end‑to‑end success and 60% higher human preference than existing simulators. Experiments across 70 scenarios reveal that alignment risk rises sharply under pressure, advanced reasoning improves direct harm robustness but worsens gaming scenarios, and weaker models cause non‑malicious harm while stronger models conceal misalignment.
By Changyi Li, Pengfei Lu, Xudong Pan, Fazl Barez, Min Yang
HAFixAgent is a history-aware program repair agent that incorporates blame-derived repository context into its repair loop. In a large-scale study on 854 Defects4J and 501 BugsInPy bugs, the agent’s best history heuristic improved repair rates by up to +64.3% with weaker models and outperformed existing agents by significant margins. The three complementary history heuristics together repair more bugs than any single heuristic, with only modest increases in agent steps or tokens.
By Yu Shi, Hao Li, Bram Adams, Ahmed E. Hassan
Sampling-Based Safe Reinforcement Learning (SBSRL) is a model‑based RL algorithm that enforces safety by jointly constraining a finite set of dynamics samples, approximating a worst‑case optimization over uncertain dynamics. It introduces an exploration strategy that limits epistemic uncertainty, removing the need for explicit exploration bonuses. The authors provide high‑probability safety guarantees and a finite‑time sample complexity bound, and demonstrate safe, efficient exploration in both simulation and real robotic hardware, with extensions to deep‑ensemble implementations for high‑dimensional continuous control.
By Luca Vignola, Bruce D. Lee, Manish Prajapat, Manuel Wendl, Melanie Zeilinger, Andreas Krause, Yarden As
The paper introduces Coverage-conditioned Group Policy Optimization (CovGPO), a method that addresses sampling imbalance in group-based policy optimization for long-horizon agentic tasks. CovGPO applies first-visit observation coverage only when all samples in a group receive zero outcome reward, giving higher relative advantages to state-exploring actions without altering the original reward signal. Experiments on ALFWorld and WebShop benchmarks demonstrate that CovGPO consistently outperforms group-based baselines, especially on hard tasks.
By Kaibing Yang, Guangfeng Cai, Shengtian Yang, Shuo He, Yu Li, Mengyi Liu, Pengwei Chen, Jun Xu, Lei Feng
TreeHCA introduces a method that couples hindsight-guided exploration with hindsight credit assignment for multi-turn large language model agents. By using changes in reference-answer likelihood to revisit intermediate histories and sample alternative continuations, it builds a tree of reasoning and tool interactions, and applies an inverse-hindsight importance correction to propagate terminal outcomes into turn-level advantages. Experiments on seven question-answering benchmarks with three Qwen backbones show that TreeHCA outperforms all baselines in aggregate exact match.
By Debajoy Mukherjee, Jeremy Carleton, Sushil Vemuri, Srinivas Shakkottai, Dileep Kalathil
This paper presents Transformer and Performer models for forecasting surface vessel trajectories in confined waters using AIS position data. It introduces positional heat maps generated via Kernel Density Estimation and encoded with a CNN to incorporate local vessel movement patterns. The models, trained with quantile loss, achieve superior performance over prior work and provide prediction intervals essential for collision avoidance.
By Frederik E. T. Sch\"oller
The paper investigates when external priors (teacher models, LLMs, or domain experts) help low‑resource structured prediction, introducing Ranking Prior Alignment (RPA) as a model‑agnostic objective that aligns a model’s ranking distribution to an external prior via a temperature‑scaled KL loss. Experiments on tree‑ensemble and multiple‑instance‑learning tasks show that a prior’s benefit depends on its quality (rank correlation with labels) and the granularity of alignment: high‑quality priors improve performance in cross‑sample tree alignment but can hurt within‑bag MIL alignment, while zero‑information priors act as entropy regularization. The authors propose a two‑factor decision rule and a protocol to separate information transfer from regularization.
By Qiye Lu, Jiang Ji, Liang Zhang
The paper introduces T‑Mem, a memory framework that uses temporal structure to manage long‑horizon multi‑party conversations. It employs Temporal Evidence Discovery to find locally concentrated evidence, Temporal Memory Structuring to model state transitions and compress low‑information spans, and Evidence Forest Linearization to serialize the evidence for answer generation. Experiments on two benchmarks show that T‑Mem outperforms existing memory frameworks across various settings.
By Junyou Guo, Chongyang Tao, Yan He, Yong Wang, Shuai Ma