WeWorkBuddy

Home / Browse

Models & research

Examine models, papers and evaluations with their methods and limitations.

800 public stories

arXiv ·

Alignment Faking Depends on the Value, Not Just the Model

arXiv:2611.03650v1 Announce Type: new Abstract: Language models can follow a directive when they believe they are monitored and abandon it when they believe they are not, a pattern termed Alignment Faking. Prior work establishes this for harmful-content objectives and a handful of moral-foundations values, leaving open how it distributes across a wider value space. We argue it is value-dependent: whether compliance survives a change in perceived oversight is a property of the model--value pairin

arXiv ·

CARD: A Proof-Carrying Certificate-to-Action Interface for Relative-Measurement Graphs

arXiv:2611.03651v1 Announce Type: new Abstract: Geometric certificates make mathematical claims about specified estimator objects, while robotic action requires a physical transform, a service decision, and timely proof support. We present CARD, a proof-carrying interface that keeps these objects distinct and records how they combine. Its lifted-fiber formulation separates semantic motion from witness repair and characterizes when projected and proof-carrying costs coincide. Its actionability re

arXiv ·

MCPGuard: Finding Conformance Gaps in MCP SDKs via Dual-Track Analysis and Dynamic Adjudication

arXiv:2611.03662v1 Announce Type: new Abstract: The Model Context Protocol (MCP) connects large language model agents to external tools and services through official SDKs that implement message validation and session logic. Specification constraints omitted by these SDKs may propagate to downstream applications, highlighting the importance of SDK-level conformance testing. Yet existing MCP conformance tools mainly exercise running endpoints and reveal little about whether SDK models enforce spec

arXiv ·

Functional-Group Knowledge Is Structured in Language Models Before It Is Spoken

arXiv:2611.03684v1 Announce Type: new Abstract: Language models serve as the reasoning core of agents for chemistry and materials science, where they must interpret molecular structure before planning syntheses or proposing materials. Yet their outputs show only what they say, not what they know. Given a prompt such as ``The primary functional group in 1-Fluorobutane (CCCCF) is'', a small general-purpose model may answer ``the alkyne group'' instead of an alkyl fluoride. The output alone cannot

arXiv ·

LOLPlay: Benchmarking Real-Time Visual Agents in League of Legends

arXiv:2611.03685v1 Announce Type: new Abstract: Visual agents need to coordinate reasoning and action while their environment evolves during inference. We introduce LOLPlay, a benchmark in native \emph{League of Legends} with 36 champion-specific tasks spanning resource management, combat, cooperation, and objectives. Agents interact through screenshots and mouse--keyboard actions; automated initialization and 82 executable subgoal checks distinguish partial progress from completion. Across 324

arXiv ·

Training LLM Agents to Automate LLM Post-Training with RSI-Dojo

arXiv:2611.03698v1 Announce Type: new Abstract: Recursive Self-Improvement (RSI) aims to enable AI systems to autonomously build and improve AI systems themselves, including models, harnesses, and underlying infrastructure. Within this broader agenda, a central capability is enabling agents to train and improve base models. However, existing studies on automated model training primarily focused on workflow- or harness-level optimizations, rather than internalizing the ability to conduct post-tra

arXiv ·

Unsatz: Evaluating Scientific Discovery Without a Given Ansatz

arXiv:2611.03713v1 Announce Type: new Abstract: Before a scientist fits any parameter, they must decide what the model contains: which quantities exist, which are hidden, how they interact, and how they are measured. Physicists call such a trial form an ansatz. Benchmarks for automated equation discovery mostly supply it, by fixing the variables, a library of candidate terms, or a law to be modified. We introduce Unsatz, a benchmark in which an LLM agent must construct the ansatz itself. Each ta

arXiv ·

Counterexample-driven autonomous research with application to excess radiated energy in a Z-pinch

arXiv:2611.03721v1 Announce Type: new Abstract: Recent breakthroughs in artificial-intelligence-assisted mathematics show that open-ended machine search can outperform human mathematicians, yet transferring this paradigm to physics requires refactoring the research workflow. We introduce Simjecture, an open-source framework in which language-model agents pursue human-defined physical conjectures. The framework tackles two central challenges in this transfer: first, because physics is empirical a

arXiv ·

EMBERS: Empirically Motivated Robustness Testing of Deep Research Systems via Defect-Guided Fuzzing

arXiv:2611.03723v1 Announce Type: new Abstract: Deep Research (DR) systems extend large language models into autonomous agents that decompose complex information needs, iteratively search the web, and synthesize evidence to answer challenging questions. However, their multi-stage execution may introduce robustness risks, as variations in user questions or intermediate search inputs may propagate across research stages and lead to incorrect answers. Existing evaluations focus on task-solving capa

arXiv ·

From Logs to Faults: Reference-Free Fault Localization for Retrieval-Augmented Generation Pipelines

arXiv:2611.03725v1 Announce Type: new Abstract: Retrieval-Augmented Generation (RAG) pipelines are widely used for domain-specific question answering, yet locating faults remains manual and time-consuming. Fault localization is challenging because RAG pipelines consist of interdependent stages, where faults can propagate downstream, masking their origin, or remain undetected even when the final answer is correct. Existing evaluation approaches often focus on individual stages and rely on referen

arXiv ·

Reasoning is topologically simple

arXiv:2611.03739v1 Announce Type: new Abstract: We formulate and investigate the hypothesis that reasoning text has simpler co-occurrence topology than non-reasoning text, and that higher-quality reasoning has simpler co-occurrence topology than lower-quality reasoning. We discuss and test applications to AI safety and AI evaluation.

arXiv ·

VagueBench: Benchmarking the Robustness of Embodied Agents on Ambient Context Illusions

arXiv:2611.03765v1 Announce Type: new Abstract: Embodied agents are increasingly deployed in open-world environments, yet current robustness evaluations remain biased toward hazardous instructions and jailbreak vulnerabilities. A more fundamental issue is largely overlooked: agents are rarely tested under semantically sufficient but contextually minimal instructions (i.e., common in real-world human communication), where auxiliary contextual cues are absent. For the first, we identify a novel ph

arXiv ·

El Agente F\'armaco: A Scalable AI Scientist for Drug Discovery

arXiv:2611.03832v1 Announce Type: new Abstract: Computational drug discovery offers a growing range of tools for proposing and evaluating molecules, but combining them and deciding what to do next still requires substantial expert effort. Here, we introduce El Agente F\'armaco, an AI scientist that uses modern large language model (LLM) technology to turn discovery goals expressed in everyday language into complex, scalable computational screening workflows. The agentic harness selects scientifi

arXiv ·

Low-Rank Hierarchical Merging for Efficient Long-to-Short Reasoning

arXiv:2611.03834v1 Announce Type: new Abstract: Long-to-Short (L2S) model merging seeks to combine the strong reasoning capability of long chain-of-thought models with the concise style of base models, but existing methods that rely on linear interpolation and first-order activation statistics fail to capture how the loss landscape responds to parameter displacement, leading to degraded accuracy on hard tasks. We propose Low-Rank Hierarchical Merging (LHM), a training-free framework that replace

arXiv ·

LaplaceDiag: Fault Diagnosis through Changes in Learned Actuator Effects

arXiv:2611.03849v1 Announce Type: new Abstract: Identifying faulty components from sensor anomalies is difficult because fault effects can propagate through a plant and different faults can produce similar observations. Detailed component models and labeled fault examples may also be unavailable. We introduce LaplaceDiag, a data-driven method that uses changes in actuator effects to narrow the search for faulty equipment. Actuators act on specific devices and mechanisms, so their changed effects

arXiv ·

Performance, Scalability and Maintenance: An Experimental Study of RPA vs. Autonomous Agents in the Healthcare Domain

arXiv:2611.03853v1 Announce Type: new Abstract: Healthcare information systems are fragmented, and retrieving data from legacy web portals is a recurrent need. Robotic Process Automation (RPA) is deterministic but sensitive to interface changes, whereas autonomous agents based on large language models (Agent Process Automation, APA) promise adaptability at a higher computational cost. Following Design Science Research, we developed and compared an RPA solution (Playwright) and an APA solution (b

arXiv ·

Predictive gains, causal losses: why interventional competence cannot be learned from data alone

arXiv:2611.03861v1 Announce Type: new Abstract: Retaining predictive gains can require persistent errors in a model's response to interventions. For deterministic predictors with finite second moments under squared-error loss, any observational advantage over the best predictor using only an outcome's direct causes requires dependence on other inputs, regardless of the learning rule. When admissible interventions vary inputs while preserving the worldly outcome, this dependence yields a positive

arXiv ·

ProofPaths: An AI Agent Framework for Assessing Student Reasoning

arXiv:2611.03882v1 Announce Type: new Abstract: Students can use large language models to complete assignments, but a correct solution does not establish whether they understand the reasoning behind it. Individualized interviews can probe that understanding but require recurring staff effort. We introduce ProofPaths, an AI agentic framework that combines rubric-based grading with individualized and interactive interviews. Multiple grading agents assess a submission against instructor guidance, e

arXiv ·

What Is This Memory Useful For? Measuring Task-Conditioned Utility in Long-Term Agent Memory

arXiv:2611.03930v1 Announce Type: new Abstract: A stored experience may answer one future question and offer little help with another. We measure this task-conditioned utility rather than assigning each memory a single importance score. A task-blind writer stores a memory without seeing the later task; a frozen reader then answers that task with and without the memory, using identical filler. A cost ledger records the resources used to form, store and expose the memory. On single-call factual-re

arXiv ·

We Ran a Research Sprint With AI. The AI Worked. The Research Didn't

arXiv:2611.03946v1 Announce Type: new Abstract: Generative AI tools for research are widely deployed, yet individual productivity gains do not guarantee better team research. We argue that how AI use is structured matters alongside access. Personal AI tools may amplify two failures under deadline pressure: a consumption bottleneck, in which generation outpaces team-visible evaluation, and AI-induced research silos, in which private human-AI loops displace shared decision-making. We motivate this

arXiv ·

Measuring Model-Specific Alignment Drift in LLM-Based Opinion Dynamics

arXiv:2611.03980v1 Announce Type: new Abstract: Large language model (LLM) agents are increasingly used as synthetic participants in opinion-dynamics experiments, often assuming that an agent's behaviour is fully determined by its prompt. We test this assumption with a framework in which LLM agents discuss a contested policy issue over repeated rounds while their stance is quantified by a fixed psychometric instrument independent of any dynamics model. Across six LLMs, committed personas are int

arXiv ·

A Guideline-Grounded Multi-Agent Framework for Personalized Exercise Prescription from Clinical and Wearable Data

arXiv:2611.03983v1 Announce Type: new Abstract: Translating population-level exercise guidance into individualized prescriptions requires integrating clinical and wearable data, including diagnoses, medications, functional limitations, and baseline activity, with condition-specific recommendations, a process difficult to scale beyond specialist rehabilitation settings. We developed FITTER, a multi-agent framework that integrates these data with condition-specific guidance to generate personalize

arXiv ·

Thinking Fast, Trading Late: System One Decision Models vs the Market

arXiv:2611.03985v1 Announce Type: new Abstract: Large language models (LLMs) read company news at scale, and backtests suggest that their readings predict stock returns. We ask how much of that predictability survives a test that rules out look-ahead, memorization and selection. We propose OLIVIA, a protocol for such tests: point-in-time inputs with company identities masked, selection on validation data only, an analysis plan frozen by cryptographic hash, and a test period unblinded once throug

arXiv ·

WASP-Mem: Let Agent Organize Procedural Memory. Human Spend Effort on Diagnosis

arXiv:2611.04003v1 Announce Type: new Abstract: LLM agents struggle to automate complex enterprise procedures because static, pre-built harness cannot capture proprietary policies, undocumented behaviors, or state-dependent recovery paths that surface only during execution. Procedural memory accrued from agent experience offers a training-free path to closing this gap -- yet unlike episodic records or semantic facts, it must consolidate repeated successes and diagnosed failures into reusable act

arXiv ·

Controlling Action-Head Co-Adaptation to VLM Representation Anisotropy

arXiv:2611.04006v1 Announce Type: new Abstract: Modern learning systems increasingly combine pretrained models with downstream heads. Despite the growing importance of such modular designs, how the downstream model co-adapts to the geometry of pretrained representations remains relatively underexplored. We study this problem through vision-language-action models (VLAs), where a pretrained vision-language model (VLM) conditions an action model for control. We find that the action model progressiv

arXiv ·

ROCm InferaSim: Cut LLM Serving Trial Costs and Deploy the Best Configuration

arXiv:2611.04032v1 Announce Type: new Abstract: Serving an LLM efficiently depends on the engine, how the model is split across GPUs, precision, scheduler settings, speculative decoding, KV-cache management, and prefill-decode disaggregation. The best configuration depends on the model, hardware, workload, and objective. Operators usually find it by trial and error. Exhaustively measuring the candidates in our study would take about 25,000 GPU-hours. ROCm InferaSim predicts serving performance f

arXiv ·

Recursa: Toward Recursively Self-Improving Foundation Model Systems

arXiv:2611.04035v1 Announce Type: new Abstract: Foundation models in changing environments can encounter unfamiliar domain knowledge that their original training did not cover. However, new domains rarely provide large amounts of expert-created training data or reliable supervisory signals for further training. We introduce Recursa, a foundation model system that adapts to new domains by learning from its own experience without human experts or a stronger teacher model. Recursa uses a knowledge