The paper introduces Dynamic Tool Output Compression (DTOC), a framework that manages context in large‑language‑model agents by storing full tool outputs in external memory and inserting compact placeholders into the active context. DTOC treats context updates as explicit, reversible operations within the agent’s reasoning loop, allowing selective reconstruction of compressed outputs when needed. Experiments on the DeepSWE benchmark show that for responsive models such as Sonnet 4.6 and GPT‑5.4, DTOC reduces input tokens and agent steps while significantly improving solve rates and lowering cost per solved task, with ablation studies confirming the importance of reversibility for maintaining performance.
By Abhay Chaturvedi, Shreya Bhattacharya, Rashmika Gopalkrishnan, Peter van der Putten
The paper argues that observing only behavior is insufficient to identify social norms in large language model (LLM) societies. It introduces an evaluation framework that also measures agents’ reported empirical and normative expectations, revealing that expectation elicitation boosts cooperation, social learning stabilizes behavior, and social selection identifies cooperators but offers limited reinforcement. The study shows that similar cooperative outcomes can stem from distinct underlying mechanisms and that expectations can be used to attribute each mechanism’s contribution.
By Rasika Muralidharan, Haewoon Kwak, Jisun An
The paper introduces GVS5H, a ledger‑based zero‑shot self‑orchestration technique that allows fresh instances of a single model to decompose coding problems and coordinate via a shared file system. On the 100 most recent hard LiveCodeBench tasks, the method improves performance by up to 25.6 points across eleven open and closed‑weight models, enabling cheaper models to reach frontier‑level accuracy. While the gains are not universal—some models see no improvement or a decline—analysis suggests that decomposition and persistent context are key contributors to the observed performance boost.
By Victor Gao (Sang Won), Vida Khosrowshahi (Sang Won), Ali Khosrowshahi (Sang Won), Xihao Sun (Sang Won), Juhyun Lee (Sang Won), Simon (Sang Won), Lee
MatchFusion is a learnable module that performs explicit-implicit instance matching for spatio‑temporal multimodal autonomous driving. It initializes pairwise affinities with geometric similarity and category consistency, then refines associations using instance embeddings to guide a residual aggregation operator for adaptive information exchange. Experiments on nuScenes show that MatchFusion improves perception accuracy, reduces FLOPs by 55.3% and GPU memory usage by 39.3%, and adds only 3.7% of total perception latency.
By Xiaoyu Li, Jiajia Fu, Long Shi, Tianyu Du, Ruihang Li, Xian Wu, Lijun Zhao, Yingtao Zhang, Lining Sun, Ruifeng Li
The paper presents a spiking neural network-based Proximal Policy Optimization (SNN‑PPO) algorithm for autonomous UAV navigation through constrained openings in civil infrastructure. By integrating spike‑based actor‑critic reinforcement learning with a stochastic Gaussian policy, the method achieved 63.77 % overall success across 3000+ episodes, improving to over 90 % in later stages and averaging 2.10 windows per episode.
By Francis Noah Walugembe, Maciej Wielgosz, Toma\v{z} Gori\v{c}an, Matej Mertik
The paper evaluates seven graph database engines, including Corvic AI, on a synthetic biomedical property graph with 1.02 million nodes and 5.34 million rows. It benchmarks query latency, bulk‑ingest throughput, point‑update latency, and correctness across a twenty‑query workload that covers neighborhood lookups, bounded paths, set intersections, anti‑joins, aggregation, ranking, temporal filters, full scans, and relational joins. The study finds that no single engine is universally fastest; performance depends on query shape, and the largest cost difference arises from bulk‑ingest throughput, which varies by three orders of magnitude and dominates total cost for workloads with fewer than about 10⁵ queries per data refresh.
By Donald Nguyen, Gurbinder Gill, Hadi Ahmadi, Christopher J. Rossbach
SAIL is a framework that transforms robot imitation learning into an iterative refinement problem, enabling test-time scaling of trajectory generation. It employs Monte Carlo Tree Search where each node represents a full trajectory and edges denote refinements, guided by an archive of successful trajectories, a vision‑language model for scoring, and step‑level feedback. Experiments on six manipulation tasks in simulation and real‑world settings show that higher test‑time compute consistently raises success rates, reaching up to 95% on complex tasks.
By Makoto Sato, Yusuke Iwasawa, Yujin Tang, So Kuroki
The paper introduces an agentic strategic asset allocation pipeline called the Self Driving Portfolio, where 44 specialized agents generate market assumptions, 21 competing methods construct portfolios, and agents critique and vote on each other's outputs. A researcher agent can propose new construction methods, while a meta agent evaluates past forecasts against realized returns and rewrites agent code and prompts to enhance future performance. The entire process is governed by an Investment Policy Statement, mirroring the document that guides human portfolio managers, thereby constraining and directing autonomous agents.
By Andrew Ang, Nazym Azimbayev, Andrey Kim
The paper reports that personal AI agents, when given users’ private data, tend to steer recommendations toward more expensive options for wealthier users across flights, health insurance, and graduate programs. In 325,000 experiments on 13 models, even when users explicitly ask for the cheapest choice, many agents still favor pricier alternatives based on inferred wealth. The effect persists when wealth is inferred from unrelated emails and can worsen when non‑financial attributes are blocked, indicating that larger models are not immune to this bias.
By Aman Priyanshu, Supriti Vijay, Brian Jabarian, Niloofar Mireshghallah
The paper introduces ChartBias, a benchmark of 820 real-world charts covering six social attributes, designed to audit bias in vision‑language models (VLMs) that interpret charts. Across 12 VLMs, the study identifies three failure modes—narrative shift, group hallucination, and preference polarity—where models produce different or misleading narratives when the referenced social group changes. A multi‑agent mitigation framework is proposed, separating evidence extraction from group‑conditioned generation and using a counterfactual judge, which reduces narrative shift while maintaining chart‑grounded reasoning.
By Mizanur Rahman, Huan Wu, Arash Asgari, Enamul Hoque Prince, Laleh Seyyed-Kalantari
CONCAT is a training‑free framework that improves the efficiency of large language model (LLM) based multi‑agent systems by clustering agents according to their initial answers and selecting cluster leaders based on confidence. It uses a Theory‑of‑Mind‑inspired heuristic to predict collaboration benefits between leaders, then prunes communications to form an ad‑hoc network that reduces latency. Experiments on three LLMs and benchmarks show up to 2.02× higher accuracy/latency ratio than LLM‑Debate and a 50.1% latency reduction on Qwen2.5‑14B‑Instruct without task‑specific training.
By Ziyang Ma, Dingyi Zhang, Sichu Liang, Jiajia Chu, Pengfei Xia, Hui Zang, Deyu Zhou
Metric-Bench introduces a new benchmark for Vision‑Language Models (VLMs) that focuses on metric‑spatial reasoning in indoor scenes by using in‑image reference objects with known dimensions. The accompanying MetricReasoner fine‑tuning recipe employs structured prompts and numerical rewards to implicitly learn 2D‑to‑3D mapping without camera intrinsics. Experiments show that this approach improves spatial metric understanding by 43.1 % over existing models and boosts downstream embodied tasks, while also delivering gains on general VLM benchmarks.
By Yuling Xi, Haokai Zhang, Muzhi Zhu, Hao Zhong, Zongze Du, Hengyu Zhao, Chenchen Jing, Yufei Yin, Bin Qin, Yongjie Yang, Zhenbo Luo, Hao Chen, Chunhua Shen
The paper "End-to-End Visual Odometry with RNNs and Attention" presents a study of deep‑learning approaches to visual odometry (VO), proposing a novel temporal attention‑based model to enhance performance. It evaluates existing end‑to‑end VO methods and explores their effectiveness on hand‑held camera data, contrasting with the typical driving‑scene training sets. The work aims to improve VO accuracy in more dynamic and complex visual environments.
By Ruiyu Li, Yinjia Liu, Alexander Yu
The paper presents a framework that builds a static point cloud prior map from past camera traversals, augmenting each point with DINOv3 semantic features. During runtime, a local prior patch is retrieved, encoded with a sparse voxel backbone, and fused with lifted multi‑view camera features in bird’s‑eye view. This fused representation is then used by sparse transformer heads to predict 3D objects and vectorized map elements, achieving improved performance on Argoverse 2 without requiring LiDAR for prior‑map construction or online inference.
By Markus K\"appeler, Rohit Mohan, Abhinav Valada
arXiv:2609.22529v1 Announce Type: new
Abstract: International law provides the normative framework through which states coordinate action, regulate armed conflict, and protect human rights, yet its t...
By Genis Skura, Roland Bouffanais, Didier Wernli
arXiv:2609.22599v1 Announce Type: new
Abstract: Large language model (LLM) judges provide a flexible and scalable method for evaluating model and agent outputs, but their verdicts can be sensitive to...
By Jackson Hassell, Farima Fatahi Bayat, Pouya Pezeshkpour, Estevam Hruschka
arXiv:2609.22682v1 Announce Type: new
Abstract: Collective intelligence depends not only on what team members know, but also on how they organize their work. When the structure of a solution is unkno...
By Aneesh Pappu, Mirac Suzgun, Yongchan Kwon, Federico Bianchi, Batu El, Mykel J. Kochenderfer, Hancheng Cao, James Zou
arXiv:2609.22987v1 Announce Type: new
Abstract: Automated operations research (OR) modeling requires LLMs to translate natural-language decision problems into correct mathematical programs. Existing...
By Ruiqing Zhao, Rui Liu, Yuan Zuo, Huarong Zhang, Xiao Han, Junjie Wu
arXiv:2609.23363v1 Announce Type: new
Abstract: Recent advances in large language models (LLMs) have led to the emergence of coding agents capable of performing complex engineering tasks, including r...
By Bowei Wang, Zhigang Fang, Zhijie Yang, Renzhi Chen, Shanshan Li, Lei Wang
arXiv:2609.23512v1 Announce Type: new
Abstract: Large language model agents are typically deployed with predefined configurations, although the required model capability, context, tools, permissions,...
By Md. Ashraful Babu