Multimodal large language models (MLLMs) can act unsafely when individually benign images and texts are interpreted jointly. Beyond implicit reasoning risk, where reasoning drifts toward harm, we identify intent-level risk, where cross-modal ambiguity silently commits the model to a hazardous premise that path-level op...
Wei Cai, Jian Zhao, Yu-Chen Yuan et al.· 0 citations
Given a state and a set of options, such as intents, tools or interface elements, System One decision models return a distribution over the options in one forward pass, without generated reasoning. Web systems increasingly build them by reading decisions from language models. Two axioms of rational choice are natural r...
Yao Long Teng, Jianguo Huang, Yanchen Deng et al.· 0 citations
Organizations increasingly encode operational procedures as multi-step agent skills, here called Agent Operating Procedures (AOPs), that must be traceable, auditable, reproducible, access-controlled, and bounded in time and cost, requirements that regulated sectors make mandatory. Yet AOPs are typically kept as coarse-...
Ali Mohammadi Shanghooshabad (JPMorgan Chase, Co.)· 0 citations
Traditional reinforcement learning (RL) algorithms face significant limitations in handling long-term tasks with sparse rewards. Recent advancements have leveraged large language models (LLMs) to enhance RL by utilizing their world knowledge for task planning and reward generation. However, planning-based approaches of...
Xiuhui Zhang, Ning Gao, Xingyu Jiang et al.· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
Behavioral consistency across repeated executions provides an informative signal for assessing the reliability of large language model (LLM) agents. However, existing approaches typically summarize consistency with a single task-level score, without identifying where behavioral differences occur or whether actions are...
Ya-Xin Ma, Wen-Bo Zhang, Satya K. Pothapragada et al.· 0 citations
Retrieval-augmented generation over text-rich knowledge graphs requires identifying evidence that contributes to answering a question, rather than merely matching its textual content. Existing agentic retrievers adaptively balance global search and multi-hop exploration but primarily rely on textual similarity to guide...
Wei-Chen Liu, Xiang-Cheng Wu, Depeng Xu et al.· 0 citations
Learning from experience requires recovering reusable knowledge and determining when it applies to a new situation. We introduce OntoAct, a hypothesis-based neuro-symbolic framework built on ontological abstraction. OntoAct turns individual experiences into axioms over entity types, relations, and conditions, making th...
CoWorld Reasoner, an agentic framework that uses a VWM as a unified visual reasoning engine, and VLM-VWM Harness Co-Evolution, which jointly revises tools, skills, and execution code based on execution feedback while keeping both model weights fixed.
Dong-Jie Cheng, Rui-Feng Yuan, Ran Wei et al.· 0 citations
RoleQ is introduced, a quantization framework that places scales by the functional roles of Transformer units rather than by the layout of the weight matrix, as conventional formats do, and leads every post-training ternary method at matched or lower memory-resident bits on perplexity, zero-shot accuracy and MMLU.
Johannes Knittel, Alec Helbling, Hanspeter Pfister· 0 citations
CIM-Embed is presented, a framework for learning instruction-conditioned multimodal embeddings for clustering, and Frozen product-trained encoders improve out-of-domain conditional grouping, while controlled transfer experiments distinguish partition transfer from instruction response.
Ying-Hua Xu, Peng-Yang Xie, Yan-Xiang Zhou et al.· 0 citations
Incremental learning (IL) aims to continually acquire new knowledge (plasticity) while retaining previously learned information (stability). However, striking a balance between plasticity and stability remains a significant challenge for intelligent systems. The human brain achieves exceptional balance, owing to variou...
Wenju Sun, Qingyong Li, Boyang Li et al.· 0 citations
Large language model (LLM) agents have demonstrated strong capabilities in complex tasks, including code generation, tool use, and multi-hop question answering. Existing self-improvement methods enhance agent performance through model optimization, harness evolution, and executable artifact refinement, while recursive...
Huan Zhang, Mingju Chen, Dongxu Zhou et al.· 0 citations
