Research
Papers and datasets worth knowing, ranked by significance and community attention.
Memento 3: Model-Based Recursive Self-Improvement through Reflective Rulebooks
We introduce Memento 3, building on the Memento series to enable frozen LLM agents to continually learn explicit world models through external memory.
From Reactive Containment to Proactive Assurance: Lessons from OpenAI, Anthropic, and Google Agent Security Incidents
In 2026, cybersecurity evaluations involving OpenAI, Anthropic, and Google agents reached real systems outside their authorized test scope.
PaperRISED: Rubrics for Agentic Multi-Environment Selection and Self-Distillation
RISED: Rubrics for Agentic Multi-Environment Selection and Self-Distillation
Embodied Turing Machines: Stateful Code for Robot Recursive Self-Improvement
We propose a different view: the embodied world is an Embodied Turing Machine, whose tape is the robot and environment state and rules are the policy.
OneSearch-VL: Unified Multimodal Deep Research Agent for Image and Video
We introduce OneSearch-VL, a unified agent centered on the Visually Grounded Evidence Graph (VGEG), which encodes these dependencies as a shared task-level reference for data construction, process supervision, and operation-level evaluation.
SuperNav: An Agentic Navigation System for Any Task in Any Scene
General-purpose service robots need navigation systems that can handle diverse human requests in unfamiliar environments, combining task generality with scene generality.
WorldGuide: Goal-Directed Video World Model for Procedural Task Execution
We formulate procedural video generation as closed-loop task execution in visual world space and introduce WorldGuide.
AgentGarten: Code Worlds for Evolving Agents
We introduce AgentGarten, a framework that couples simulators and game engines with a shared neural renderer to build real-time interactive environments.
Accurate but Not Humble: Evaluating Epistemic Humility in LLM Agents under Knowledge Conflict
We propose to evaluate agents on epistemic humility (EH): the agent's willingness to recognize, act on, and communicate uncertainty during task execution.
Self-Retrospection Distillation: Turning Post-hoc Experiences into Prior Foresight
Reinforcement learning with verifiable rewards (RLVR) turns agent experience into learning signals primarily through scalar outcome rewards after interaction.
CADFather Reconstructs Parametric CAD Programs Using Coordinated Tools
CADFather is an autonomous agentic system that coordinates complementary tools and a vision-language assistant to reconstruct parametric CAD models from 3D meshes without additional training.
Ecology of AI Agents: Collaboration Creates a Population Threshold for Takeoff
Here, we develop an ecological theory of AI-agent populations based on a population growth equation in which fitness (growth rate) depends on cybersecurity capability.
Can AI Agents Make Open-Ended Scientific Discovery? Evidence from Station
Recent AI systems have made rapid progress in scientific discovery when given well-defined metrics, but whether they can autonomously undertake open-ended scientific discovery remains unclear.
Forms of LLM-Integrated Applications from LLM-Chats to Autonomous AI Agent System
Large language models (LLMs) are increasingly embedded as components in software systems, marketed under labels such as chatbot, copilot, retrieval-augmented generation, workflow, coding agent and AI agent.
Skill Constellations: Tracing the Supply Chain of Agent Skills on GitHub
Agent skills are SKILL.md instructions and scripts that AI coding agents such as Claude Code and Codex run with the permissions of their user.
MIRA: A Musical Intent Refinement Agent for Aligning Text-to-Music Generation with User Intent
To bridge this gap, we formulate text-to-music intent alignment as satisfying a per-request rubric of independently verifiable items covering both a request's explicit requirements and its implied musical intent.
Agent Plasticity: Measuring Self-Improvement Through Experience
We introduce agent plasticity, the efficiency with which an agent converts experience into gains in future held-out performance.
RFChipAgent: Multi-Agentic AI Flow for Analog/RF Chip Design
We present RFChipAgent, a first-of-its-kind multi-agent flow of large language model (LLM) agents for end-to-end analog/RF circuit design automation, in which AI agents collaboratively orchestrate the complete design flow under human supervision.
AgentHorizon: Evaluating Agentic Judges for Long-Horizon Computer-Use Tasks
To this end, we introduce AgentHorizon, a benchmark of 1,373 computer-use tasks (instruction-trajectory pairs) drawn from 166 hours of human-recorded trajectories spanning three operating systems.
Agentic AI-Assisted Modeling for Production Scheduling: Assessment in Constraint Programming
Developing optimization models for production scheduling requires substantial expert effort.
One Word Opens the Gate: The Option-Channel Attack on Typed Decision Models as Agent Guardrails
A typed decision model reads a piece of text and returns a probability over caller-defined options, each with a short written definition, generating no text.
Mental-Models for Multi-Agent Systems
We introduce mental-model-enabled agents, a framework that equips an agent with a latent mental model of its counterpart, allowing it to infer hidden beliefs, intentions, and likely reactions from the observed history and use these inferences to guide action selection.
Why LLM Agents Favor Their Group: Stakes, Observed Norms, and Reputation
Language-model agents favor their own group because they have watched their members favor each other.
AgentEvolver: System-Wide Self-Evolution Through Task Execution
We present AgentEvolver, a system for developing capabilities during task execution while keeping the foundation model fixed.
Shared and structured inputs undermine collective random choice by reasoning AI agents
Random selection is widely used in resource allocation and auditing, making reliable implementation essential for AI-agent systems.
Cross-Provider Review as a Runtime Contract for Coding Agents: A Controlled Pilot and Fault-Injection Study
We describe an advisory cross-provider review contract: distinct resource pools, bounded execution, restricted reviewer capabilities, complete input delivery, usable semantic output, explicit failure states and durable per-attempt evidence.
A self-learning scientific agent for X-ray diffraction
Here we introduce Gan Jiang, a self-learning agent for powder X-ray diffraction built on a diffraction-analysis ecosystem we developed: XMatcher, XQueryer, XDecomposer and WPEM.
Safe, Persistent, and Evolving Agent Harness for Understanding Partially Observable Worlds
To address these challenges, we introduce E-Ledger, a multi-agent harness for safe and persistent execution.
Code Understanding is a Bottleneck for Coding Agents
We present CABRA: a Coding Ability Blueprint for Rigorous Agent evaluation.
GeoNatureAgent (GNA): A Framework and Benchmark for Pre-Production Evaluation of Tool-Using Agents on Geospatial and Environmental Tasks
We introduce GeoNatureAgent (GNA), a framework for pre-production evaluation of tool-using agents: a fixed sixteen-tool geospatial interface published as a Model Context Protocol (MCP) server, so the agent under test is the only variable, scored against an identical tool layer, task suite, and deterministic scorer.
I would rather quit NLP than read another paper like this: The rise of antithesis in NLP papers
We study the construction rather than in ACL papers from 2019, ACL-style arXiv papers from 2026, and papers written by GPT models from the same titles and abstracts.
AgentTime: Can Agents Estimate and Control Their Own Runtime?
We present AgentTime, a benchmark for testing whether agents can work for a requested duration, predict their runtime, and estimate elapsed time afterward.
Multi-Aspect Runtime Verification for Simulation-Based V&V of LLM-Enabled Autonomous Agents
We present a multi-aspect runtime-verification framework that decomposes a natural-language policy clause into a typed spatial/temporal/semantic triple over one canonical event stream, checks each aspect with its own monitoring specification, and fuses the verdicts through a four-valued algebra that carries provenance.
How Do Agentic LLMs Decide to Call Tools? A Tool-Call Vector Shaped by Suppression
To obtain such a variable, we propose a method that converts complex agentic prompts into minimal contrastive pairs in which a single request verb determines the tool-call decision: replacing an execution-verb (e.g., write) with an analysis-verb (e.g., discuss) reliably flips the decision, suggesting it is mediated by a compact internal state.
RunningTab: Direct Workspace Interaction with Environment-Side Tabs
To address this, we present RunningTab, a framework that equips direct workspace interaction with an environment-side tab: a per-task record of what the task still owes, kept by the environment alongside the agent.
Trajectory-Guided Fault Localization for Agent Skill Evolution
To address this gap, we propose SkillMorph, a skill-evolution approach based on trajectory-guided fault localization in agent skills.
Loud Failures, Quiet Failures: Fault Detection and Recovery in Tool-Using Language Model Agents
Prior work has shown that language models over-trust tool outputs that fail silently; we ask how that over-trust plays out across the stages of failure handling in multi-turn agents.
Narrow and Deep: An Ontology Tower as the Knowledge of an LLM Agent for an Industrial Equipment System
This study proposes the ontology tower, a narrow-and-deep ontology of a single equipment system whose knowledge deepens in two ways: through quantities derived from the measured points by physical relations, and through lessons from the operating journal incorporated as knowledge nodes.
What Output-Only Review Cannot Verify: Study Contracts for Research Agents
We propose study contracts that bind declared experimental choices, run obligations and claim scope to recorded execution evidence, and distinguish this contract-relative verification from scientific truth.
DeltaReplay: Task-Relative Memory Reuse for Mobile GUI Agents
Memory-augmented mobile GUI agents store successful execution trajectories and reuse them in later tasks, but a stored trajectory rarely matches a new task exactly.
One Skill Too Many: How Co-Installed Skills Conflict in Coding Agents
Coding agents are extended with agent skills, directories whose SKILL.md tells the model when and how to perform a task.
Coding-Agent Benchmarks Should Match Their Users' Task Flows
We present SWE-TaskFlow, an approach for transforming any issue-derived benchmark: it preserves the verified tasks and tests while steering the interaction toward a target Task Flow through prompt splitting and verifiable repository QA, with a TaskFlow Alignment Score (TFAS) for selecting among generated trajectories.
LiveMACE: Process-Aware Evaluation of LLM Agent Capabilities in Evolving Markets
We introduce LiveMACEBench, a process-aware benchmark that uses live financial markets as a naturally evolving testbed for persistent LLM agents.
Runnable Commit Untangling for Coding Agents
Commit untangling offers the promise of organizing such large patches into untangled, manageable commits.
Ream: Unfolding Mutual Awareness in Human-Agent Workspaces
We present Ream, a literature review workspace that supports mutual awareness through structured artifacts, bidirectional engagement tracking, and localized visualizations.
GROB: A Multi-Agent Architecture for Public-Trace Investigation of Candidate Agentic Activity
We present GROB, a multi-agent architecture for investigating candidate autonomous-agent activity through public Internet traces when privileged telemetry is unavailable.
Who Verifies the Verifier? Co-Evolving Inspectable Graders with Self-Improving Agents
We make the verifier the evolving object: an inspectable expression over small, mostly deterministic drawback detectors, synthesized from clustered failures, gated at birth, and selected for agreement with a ten-item anchored reference set plus consensus over unlabeled outputs, never for the agent's score.
When the Governor Becomes the Disturbance: Control-Generated Disturbance and Cost-Aware Backoff in Governed Tool-Using Agents
We study this possibility in a controlled file-recovery environment where increases in regulatory intensity trigger experimentally imposed tool failures.