The paper investigates how adversarial signals can infiltrate large‑language‑model (LLM) based multi‑agent trading systems through the agents’ communication channels. By restricting the attacker to realistic inputs—source data and prompts—it studies role‑specific attacks on four functional roles (Analyst, Researcher, Trader, Risk Manager) and evaluates four communication topologies under data‑ and agent‑level attacks. Experiments across multiple assets, backbones, and target directions show that no architecture is inherently robust, highlighting the need for safer designs in agentic trading systems.
By CheolWon Na, Hao Ni, Lukasz Szpruch, Zhangyang Wang, Dhagash Mehta, Saurabh Nagrecha, Alejandro Lopez-Lira, Chanyeol Choi, Yongjae Lee, Jee-Hyong Lee
SPO++ is a new reinforcement learning method that eliminates the need for waiting on sibling rollouts by using a persistent prompt-level value estimate. It corrects a mismatch in the original SPO approach by standardizing terminal-outcome advantages under an action-token measure and reorganizes prompt evidence based on the policy event that generated it. Experiments on ALFWorld and Math‑TIR show that SPO++ improves online learning efficiency, with action-token-measure normalization identified as the key component.
By Kai Ruan, Jinghao Lin, Qianshan Wei, Ziqi Zhou, Zihe Huang
Recursive self-improvement (RSI) remains hard in long-horizon tasks, where growing histories obscure the task state and misalign skill invocation. We introduce Recuris, a recursive Experiential-Workin...
Web agents that act from rendered pixels avoid the fragility and heavy token cost of reading a page's HTML or accessibility tree, but training them depends on large amounts of high-quality interaction...
Scientific agents increasingly analyze data, execute code, and produce research artifacts, yet most benchmarks emphasize final answers, isolated programs, or a single domain. We introduce FrontierChal...
Outcome-supervised search agents learn when and how to retrieve evidence, but terminal rewards neither localize intermediate errors nor redirect an ongoing trajectory before those errors compound. Tre...
Large language model (LLM) agents are trained with reinforcement learning (RL) for complex decision-making tasks. However, most RL-trained agents remain episodic and cannot accumulate reusable knowled...
Existing linear program (LP) and semidefinite program (SDP) relaxations for rectified linear unit (ReLU) neural network (NN) verification yield overly-conservative safety guarantees due to significant...
Agentic systems increasingly gate actions on a model's own stated confidence, which assumes confidence tracks correctness at the moment of acting. We test this in a hidden-information chess variant wh...
Large Language Model (LLM) agents increasingly solve long-horizon tasks through multi-turn interactions with users and external tools. In these settings, relevant task information often unfolds over t...
Human collective intelligence depends on transmission processes: who shares what with whom, how, and when. While these processes emerge from individual cognition, they can also be directed by delibera...
Visual Language Models (VLMs) excel at describing visible scene content but struggle to reason about dynamic multi-agent interactions, where action semantics depend on coordinated roles and spatial-te...
Safeguarding language model agents requires assessing complete execution trajectories under context-dependent safety policies. Existing policy-aware safeguards mainly rely on prompting or supervised f...
Coding agents re-send large file reads and tool outputs to a frontier LLM every turn, and this context dominates their token bill. General-purpose prompt compressors are trained on prose and suit code...
The majority of our everyday activities are procedural and consist of sequences of interdependent steps. However, existing benchmarks for Visual Agents and Visual Language Models (VLMs) overlook the e...
Telephone fraud is pervasive and costly, but its inner workings are rarely observed at scale. We analyze a complete corpus of 10,211 inbound scam and spam calls -- 913 hours of audio and 330,956 trans...
Commercial design platforms increasingly edit documents through large language model (LLM) agents, but two practical problems block reliable deployment: legacy document formats expose only \emph{flat}...
Universal multimodal embeddings are becoming a core component of modern AI systems, enabling heterogeneous content to be represented in a shared space for applications such as retrieval, recommendatio...
The paper introduces Clinical Graph-JEPA, a framework for building and refining predictive patient-state knowledge graphs from clinical records. It combines multi-agent relation proposal, ontology-aware normalization, deterministic evidence scoring, and JEPA-based latent refinement to construct evidence-scored graphs from MIMIC-IV data and recover missing clinical relations. Experiments show that injecting discharge-note representations into note-grounded entities boosts leave-one-out MRR by 31% relative improvement.
By Kushagra Yadav, Nalin Prabhath, Amit Lamba, Goeun Han, Yining Mao
arXiv:2608.23330v1 Announce Type: new
Abstract: Video understanding requires intelligent agents to transcend mere recognition of visual facts and comprehend the underlying intents behind human action...
By Jiapeng Li, Ping Wei, Wenjuan Han, Song-Chun Zhu, Lifeng Fan