Flagged for: agent, autonomous, autonomy, alignment
In recent years, artificial intelligence has made extraordinary progress thanks to large-scale models capable of generalization and the generation of complex outputs. However, transferring this potent...
Flagged for: agi, reasoning, llm, language model
On-Policy Self-Distillation (OPSD) has emerged as a popular paradigm for large language model (LLM) self-improvement, allowing models to act as their own teachers by leveraging privileged information ...
Flagged for: agent, multi-agent, reasoning, llm
This paper presents SurgicalRoomAgent, a voice-interactive multi-agent system for smart operating rooms based on large language models (LLMs). The system achieves natural language understanding, devic...
Flagged for: autonomous, reasoning, chain-of-thought, interpret
Chain-of-Thought (CoT) prompting enables LLMs to perform explicit, step-by-step reasoning, creating opportunities for sophisticated autonomous robots. However, recent research reveals that reasoning m...
Flagged for: agent, agentic, autonomous, reasoning
Scientific workflow management (WMSs) systems automate execution, yet orchestrate using fixed, hand-tuned rules. LLM agents promise more autonomous orchestration, but it remains unclear where to intro...
Flagged for: agent, multi-agent, reasoning, interpret
Inferring speaker relationships from spoken conversations is an important step towards socially aware speech understanding. However, this task remains underexplored, and supervised modeling is costly ...
Flagged for: recursive self-improvement, agent, autonomous, alignment
While research on recursive self-improvement (RSI) has predominantly automated model training pipelines, reliable autonomous development demands a missing pillar: post-hoc monitoring and auditing to u...
Flagged for: agent, autonomous, planning, llm
Ensuring the safety of autonomous driving is a critical challenge. Scenario-based testing is a systematic process used to validate Autonomous Driving Systems (ADSs), but it remains a fragmented modula...
Flagged for: agent, multi-agent, agentic, autonomous
Rapid bespoke commissioning of the Cognitive Digital Twin (CDT) is a major challenge in reconfigurable manufacturing. Traditional digital twin (DT) construction methods primarily focus on geometric re...
Flagged for: agent, multi-agent, reasoning, llm
Large language models (LLMs) show potential for medical tasks, but their single-turn question-answer format does not reflect how clinical diagnosis is performed in practice. As a result, they remain l...
Flagged for: agi, reasoning, llm, language model
Hallucination-where a language model generates outputs that are factually incorrect or unsupported by the source-is a major challenge for both prompted and fine-tuned language models. Detecting halluc...
Flagged for: agent, multi-agent, tool use, reasoning
Computer-aided engineering (CAE) simulation is among the largest and most demanding areas of engineering, where setting up a solver such as OpenFOAM, FEniCS, or COMSOL takes real expertise. Large lang...
Flagged for: agi, agent, multi-agent, autonomous
Multi-agent AI science ecosystems rely on agents possessing tools that allow them to communicate, coordinate, and build on each other's work. Yet this shared infrastructure can also introduce vulnerab...
Flagged for: agent, agentic, autonomous, reasoning
Large language model (LLM) agents are increasingly proposed as autonomous SOC analysts, but two limitations make them unreliable at enterprise scale: a finite context window cannot hold a multi-thousa...
Flagged for: world model, foundation model, agent, autonomous
Knowledge graphs describe reality in crisp assertions, while the systems now consuming them, foundation models and autonomous agents, reason natively in probabilities. We argue that this mismatch is w...
Flagged for: agi, agent, tool use, reasoning
Large Language Model (LLM) agents based on the ReAct paradigm have demonstrated remarkable capabilities in tool use and task execution. However, ReAct suffers from a fundamental efficiency problem: ev...
Flagged for: agi, alignment, llm, language model
Benign fine-tuning severely weakens the safety alignment of large language models (LLMs), so we study why refusal behavior is so fragile. While prior work often attributes this failure to gradient con...
Flagged for: agi, agent, multi-agent, reasoning
Medical referral (directing patients to the appropriate hospital department) is a complex decision-making process requiring the synthesis of multimodal data, including patient narratives, laboratory i...
Flagged for: agent, agentic, reasoning, interpret
Spectroscopic structure elucidation is central to molecular analysis, but recent Large Language Model (LLM)-based methods mostly formulate it as direct spectrum-to-SMILES generation. Although this par...
Flagged for: superintelligence, agent, agentic, reasoning
Recent advances in large reasoning models (LRMs) have shown that reinforcement learning with verifiable rewards (RLVR) can substantially improve reasoning in mathematics and code, where outcomes can b...
Flagged for: agent, reasoning, embodied, robot
Embodied navigation requires agents to translate heterogeneous goals and visual observations into actions across tasks, environments, and robot embodiments. Modern vision-language models (VLMs) alread...
Flagged for: foundation model, agent, agentic, reasoning
Long-horizon physical-world agents must reason over distant goals while grounding decisions in reliable closed-loop behavior. Today's foundation models split these capabilities: vision-language models...
Flagged for: agent, agentic, reasoning, llm
Retrieval-Augmented Generation (RAG) improves the factuality of large language models (LLMs), yet existing RAG systems often struggle with complex, multi-step reasoning that requires adaptive retrieva...
Flagged for: agent, multi-agent, reasoning, llm
Large language model (LLM)-based agents have shown strong potential in solving complex tasks through multi-step reasoning, yet they remain vulnerable to execution failures. Accurate failure attributio...
Flagged for: agent, agentic, autonomous, llm
Autonomous large language model (LLM) agents increasingly face reliability, context consumption, and execution stability bottlenecks when deployed on complex, long-horizon tasks. While monolithic prom...
Flagged for: agent, multi-agent, planning, embodied
Agent systems powered by multimodal large language models (MLLMs) have advanced rapidly in recent years, yet existing embodied-agent benchmarks still lack fine-grained diagnostics for multi-agent coor...
Flagged for: agi, agent, agentic, llm
Reliable handling of unanswerable questions (UAQs) is critical for trustworthy LLM-based agents. Although memory is widely used in agent systems, its role in reliable UAQ handling remains unclear. We ...
Flagged for: agi, agent, multi-agent, agentic
Despite their potential in standardized graph tasks, Large Language Models (LLMs) remain brittle to real-world shifts in node identifiers and task formulation. While deterministic graph tools are inva...
Flagged for: agent, agentic, autonomous, reasoning
As large language models (LLMs) are deployed as autonomous agents, safety failures increasingly involve consequential actions. We study agentic misalignment, where agents take harmful actions under go...
Flagged for: agi, planning, interpret, robot
Effective human-robot collaboration in industrial settings requires robots to understand human intentions and assist with task planning, reducing workload. Recent works have explored the use of Multi-...
Flagged for: agent, agentic, tool use, interpretability
LLM-based agents are increasingly deployed in product-level execution harnesses, where jailbreaks can trigger harmful tool use and persistent state changes, creating greater risks than unsafe text gen...
Flagged for: agent, agentic, reasoning, interpret
Large Language Models (LLMs) are evolving from performing end-to-end mathematical reasoning to integrating agentic intelligence. However, most existing math benchmarks evaluate only final answers. Thi...
Flagged for: agi, world model, agent, robot
State-of-the-art action-conditioned video models are typically restricted to a single robot embodiment, preventing them from leveraging the vast corpus of heterogeneous video data that contains rich s...
Flagged for: agi, agent, multi-agent, llm
As large language model (LLM)-based multi-agent systems (MASs) are increasingly applied to long-horizon complex tasks, their reliability has emerged as the core bottleneck hindering their real-world d...
Flagged for: foundation model, agent, agentic, planning
Agent capability is not determined by the model alone. The agent harness, encompassing memory management, planning strategy, action protocol, and tool/skill orchestration, can dominate the contributio...
Flagged for: self-improving, agent, multi-agent, llm
Large language models (LLMs) have enabled AI scientist systems to automate scientific discovery, yet existing approaches most rely on static prompting or fixed workflows and fail to accumulate experie...
Flagged for: agent, multi-agent, reasoning, llm
Multi-agent large language model (LLM) workflows have emerged as a powerful paradigm for solving complex, open-ended tasks through collaborative reasoning among specialized LLM agents, but they incur ...
Flagged for: agent, multi-agent, agentic, language model
Despite their strong ability to generate code, large language models often fail to produce secure code, as their outputs frequently contain security vulnerabilities. Secure code generation is inherent...
Flagged for: agent, multi-agent, planning, reasoning
Manufacturing process planning transforms heterogeneous design information into coherent manufacturing decisions. However, existing approaches focus on isolated subtasks, such as feature recognition, ...
Flagged for: recursive self-improvement, self-improving, agent, llm
Self-improving LLM agents refine answers, not the process that produces those answers. Systems that add a meta-level hold that level fixed, and those that edit themselves must leave part of their own ...
Flagged for: agent, agentic, tool use, llm
Clinical diagnosis is an active evidence-seeking process in which clinicians acquire evidence, update competing hypotheses, and decide when the available evidence is sufficient for diagnosis. Yet many...
Flagged for: agent, agentic, planning, interpret
Agentic AI frameworks interpret open-ended task goals and decompose them into multi-step plans. Richer information about embodiment-specific capabilities, physical preconditions, and cross-robot coord...
Flagged for: agent, multi-agent, tool use, planning
Large language models (LLMs) have shown strong capabilities in reasoning, planning, and tool use, but many scientific and engineering tasks require more than plausible text and code generation. They r...
Flagged for: agent, agentic, alignment, llm
Agentic large language model (LLM) systems are reshaping scientific workflows in chemistry and drug discovery, but evaluating their open-ended, tool-augmented outputs remains a fundamental bottleneck....
Flagged for: agent, agentic, autonomous, llm
As large language model (LLM) agents move from conversation to executing code, reading local files, and orchestrating external tools, a single agent hijacked by a malicious third-party skill can cause...
Flagged for: agent, agentic, tool use, tool-use
Mid-training is increasingly recognized as a critical stage for shaping the capabilities of large language models. Recent work has shown that targeted mid-training can strengthen reasoning-intensive a...
Flagged for: agent, multi-agent, autonomous, reasoning
Autonomous vehicles require robust perception and decision-making capabilities to operate in diverse and unseen scenarios. While reinforcement learning and rule-based methods can provide effective con...
Flagged for: agent, multi-agent, agentic, reasoning
Large language model (LLM)-based multi-agent systems (MAS) often exhibit complex failure modes, which frequently cause agents to produce incorrect outcomes. This motivates the task of Agent Failure At...
Flagged for: agi, autonomy, llm, language model
As large language models (LLMs) are granted increasing autonomy, it is essential to investigate methods that can induce unsafe behavior. We propose a novel white-box attack inspired by locate-then-edi...
Flagged for: agent, agentic, chain of thought, interpretability
Many areas of AI research, such as language model interpretability and chain of thought faithfulness, seek to explain model behaviors. But what constitutes a "good" explanation? In this work, we evalu...
Flagged for: agi, agent, multi-agent, autonomous
Achieving human-level competitive intelligence and physical agility in humanoid robots remains a profound challenge, particularly in contact-rich and highly dynamic tasks such as boxing. While Multi-A...
Flagged for: agent, agentic, planning, ai safety
Job Safety Analysis (JSA) and pre-task planning can benefit from prior incident records, yet historical accident data is often stored as unstructured narratives that are difficult to consult at the po...
Flagged for: agent, reasoning, embodied, robot
Large Language Models (LLMs) have demonstrated capabilities in in-context learning, task decomposition, step-by-step reasoning, and code generation, driving their gradual evolution from text generatio...
Flagged for: agi, agent, reasoning, llm
Large Language Models (LLMs) have achieved remarkable progress in code generation, yet ensuring correctness in complex, repository-level tasks remains challenging. Existing approaches often use genera...
Flagged for: agent, agentic, autonomous, llm
Large Language Models (LLMs) are increasingly being deployed in cybersecurity operations to assist cybersecurity analysts with rapid decision-making against emerging threats. However, there is a main ...
Flagged for: agent, multi-agent, agentic, scaling
Multi-modal retrieval-augmented generation (RAG) is a key technique for visually rich long document understanding. Existing multi-modal RAG methods are progressively advancing toward multi-agent syste...
Flagged for: agi, agent, agentic, llm
As agentic tasks grow in complexity, LLM agents increasingly rely on experiential memory to reuse procedural knowledge across tasks. Effective memory design must jointly address what to store, how mem...
Flagged for: agent, agentic, planning, reasoning
Vision-language Models (VLMs) excel at 2D grounding, spatial reasoning and agentic tool-based planning in static scenes. However, consider asking a home robot "Is my medication still in the cabinet?" ...
Flagged for: agent, agentic, autonomous, tool use
Large language model (LLM) agents are evolving from conversational assistants into autonomous systems that execute long-horizon tasks through reasoning, tool use, code generation, and workspace manipu...
Flagged for: agent, multi-agent, alignment, llm
Large language model based multi-agent systems usually communicate in text, i.e., using discrete tokens. However, text introduces a discrete bottleneck. Converting the sender's continuous hidden state...