arXiv subjects

Jesse Thomason

Publications and source records attributed to Jesse Thomason.

At least 19 recordsLinked to original sources

arXiv2026

Algorithmic Primitives of Theory of Mind Benchmarks

Do Theory of Mind (ToM) benchmarks really test theory of mind, or are they actually testing parsing and algorithmic state tracking? While benchmarks vary modality and dialogue length, this paper finds that they re-encode the same few state transition algorithms. We create PDDL-Mind, a neuro-symbolic approach that recovers such algorithms and upstream inductive biases of ToM benchmarks explicitly by parsing narratives into the Planning Domain Definition Language (PDDL) and executes candidate actions against a predefined set of update algorithms. Doing so provides program-verified state traces to the model. We use our method's performance gain to audit how much of a benchmark's difficulty our inductive bias absorbs. Across MMToM, MuMA, and FanToM, outsourcing parsing to programs alone brings performance close to strong agentic pipelines, while outsourcing state tracking to PDDL moves GPT-4o to close-to-human performance. We demonstrate that complex ToM benchmarks primarily punishes Large Language Models' unreliable implicit implementation of state tracking and parsing algorithms rather than limitations in belief and counterfactual reasoning.

cs.CL↗

arXiv2026

From Relevant Memories to Sufficient Evidence: Rethinking Memory Utilization in LLM Agents

Long-term memory enables LLM agents to reuse past experiences, yet retrieving relevant memories does not necessarily provide sufficient evidence for answering a query. Essential information may remain fragmented across memory records, obscured by abstraction, or disconnected from the relations required for reasoning. We identify this relevance-sufficiency gap and propose QEMem, a framework that reformulates memory utilization as the construction of query-specific evidence sets, shifting the focus from individual retrieval relevance to collective evidential support. QEMem integrates three complementary mechanisms: query-guided discovery retrieves evidence across multiple memory granularities; source-grounded reification organizes it into structured ViewCards with explicit evidential roles and source references; and sufficiency-guided refinement selectively recovers missing information. Operating on existing memory stores without modifying memory formation, QEMem achieves 91.43% LLM-judge accuracy on LoCoMo with Claude Opus 4.5, outperforming the strongest evaluated baseline, MemR$^3$, by 7.6 percentage points and NEMORI by 10.7 points under a shared memory store. Further analyses demonstrate improved evidence coverage and strong performance on questions requiring cross-session evidence integration.

cs.AI↗

arXiv2026

SynCo: Data Synthesis Co-Training for Self-Evolving LLMs via Multi-Agent Reinforcement Learning

Self-evolving LLM agents promise to improve autonomously through continual interaction and learning, reducing their dependence on manually curated supervision. Realizing this promise requires not only updating the agent, but also evolving its training experience as its capabilities change. However, most existing pipelines rely on static datasets or separately updated synthesis models, causing previously useful tasks to become trivial while overly difficult tasks remain uninformative. This growing mismatch between agent capability and training experience limits sustained self-improvement. To address this problem, we propose SynCo, an agentic data synthesis co-training framework for self-evolving LLMs based on multi-agent reinforcement learning. SynCo jointly optimizes two independently parameterized agents: a Synthesizer that constructs training tasks from the Reasoner's evolving capability state, and a Reasoner that learns from the resulting experience. Each synthesized task induces multiple Reasoner rollouts whose outcomes provide complementary rewards to both agents. Correctness feedback improves the Reasoner, while task quality, answer reliability, and outcome-grounded teachability guide the Synthesizer. Their updates are fed back into subsequent synthesis rounds, allowing the task-solving policy and its training distribution to evolve together. Extensive experiments across eight mathematical reasoning benchmarks demonstrate that SynCo substantially outperforms a broad range of existing synthetic-data methods and controlled baselines, achieving the strongest overall performance while deriving most of its gains from previously unsolved problems.

cs.AI↗

arXiv2026

Towards an Extensible Benchmark for Spoken Dialogue with Social Robots

Language models provide a plug-and-play interface between humans and robots, but important challenges remain when speech, dialogue, fast interaction, and collaboration are required. We propose a benchmark for the community to use as a way to explore common spoken dialogue artifacts between robots and humans, including requests for clarification, interruptions, embodied signals (e.g., head nods or facial cues), and time constraints. We also explain our vision to extend the benchmark for other aspects of human-robot interaction that are important to the larger research community. To facilitate the benchmark, we further propose using \textit{Retico}, a real-time communication framework that fulfills important technical requirements to enable robots to have spoken dialogue capabilities.

cs.RO↗

arXiv2026

SWAP: Stepwise Action Policy Routing for Vision-Language-Action Models

Robot manipulation systems using Vision-Language-Action (VLA) model backbones typically use just one VLA for task execution. However, individual VLAs do not perform well across different task states and environments. We introduce a framework for dynamically composing multiple VLA policies during execution: StepWise Action Policy Routing (SWAP). SWAP formulates policy routing as an offline reinforcement learning problem, learning a routing critic that selects the most appropriate policy at each decision step given the current observation. SWAP enables robots to select new policies to execute online rather than committing to a single policy for the duration of an episode. We evaluate SWAP on both real-world DROID manipulation tasks and LIBERO simulation experiments. SWAP improves over fixed-policy execution and routing baselines, giving absolute improvements in real-world task success up to 33% while reducing successful trajectory robot action step length by 28.3%.

cs.RO↗

arXiv2026

Skeleton-and-Strategy Prompting: Training-Free Negation Understanding for Vision-Language Models

Despite the strong performance of Vision-Language Models (VLMs) on a wide range of visual question answering (VQA) tasks, these models consistently struggle to understand negation and produce incorrect answers when questions involve negated clauses. To address this limitation, we propose Skeleton-and-Strategy Prompting (\textbf{SSP}), a training-free, in-context learning method that improves VLM negation understanding capabilities without any parameter updates. Given a negation question, our method first abstracts the underlying question structure into a skeleton, retrieves a small set of same-skeleton questions from a lightweight question pool, then prompts the VLM to analyze their shared negation pattern and synthesize a single-sentence answering strategy. The skeleton and strategy are prepended to the test sample to guide the model correctly tackle the negation problems. Experiments on multiple negation VQA benchmarks show that SSP achieves state-of-the-art performance on negation-focused VQA tasks while remaining computationally efficient.

cs.CV↗

arXiv2026

Colosseum V2: Benchmarking Generalization for Vision-Language-Action Models

Vision-Language-Action (VLA) models demonstrate promising generalization in robotic manipulation, driven by advances in large-scale vision and language pre-training. This progress can be misleading. Despite the zero-shot perception and language capabilities of VLAs, their overall task performance often degrades under distribution shifts, revealing gaps in how these systems translate high-level understanding into robust behavior. To systematically study this gap, we introduce Colosseum V2, a large-scale simulation benchmark for evaluating VLA generalization in robot learning across diverse conditions. The benchmark comprises 28 tasks spanning 13 task categories and two robot morphologies, covering a wide range of manipulation primitives and long-horizon behaviors. Built on the ManiSkill simulator, Colosseum V2 enables fast, GPU-parallelized evaluation and supports both in-domain and out-of-domain testing at scale. We evaluate state-of-the-art methods, including Action Chunking Transformers (ACT) and Pi0.5, and reveal limitations in both base performance and generalization. We demonstrate strong correlations between simulation and real-world metrics that support the ecological validity of the benchmark. By standardizing tasks, metrics, and evaluation protocols within a unified benchmark, Colosseum V2 enables reproducible and fair comparisons, reduced evaluation overhead, and accelerated progress toward general-purpose robot policies.

cs.RO↗

arXiv2026

Toward Interactive Understanding of Code APIs

Empowered by advances in Language Model agents, systems have made substantial strides in code generation and understanding. However, these approaches often rely on read access to the relevant code, an assumption which does not hold when dealing with external APIs. In this work, we introduce the PAU (Python API Understanding) benchmark, where we provide models with black-box, API-level access to code snippets. Models must query the API with exploratory inputs and draw insights from the resulting outputs, with the goal of describing the snippet's true functionality. By treating the code snippets as external tools that must be understood via interaction alone, PAU studies the more general problem of unsupervised tool understanding, specifically for tools implemented as Python methods. Despite recent progress in coding agents, even frontier models struggle to achieve high performance on PAU, with the best model (Claude-4-Opus) failing to understand over 45% of the PAU test set. An investigation into the common error modes reveals that models are overconfident; they often overrate the quality of their current hypothesis, leading to insufficient exploration and premature termination. Finally, we take inspiration from the Asymmetric Actor Critic (AAC) paradigm, frequently used in robot learning, to post-train models for interactive code understanding. Models trained with AAC conduct more active exploration of the APIs, with an AAC-tuned Qwen3-8B model matching the performance of GPT-5-mini.

cs.AI↗

arXiv2026

GameBoyWorlds: A Testbed for Self-Improvement in Embodied Video Games

Powered by expert guidance, agents can operate in interactive environments; however, it is unclear whether they can learn autonomously from their own experience. To evaluate such self-improvement methods, we introduce GameBoyWorlds, a testbed for agentic self-improvement in video games. GameBoyWorlds-Execution evaluates task execution on a collection of 5 distinct game series. Agents are allowed access to dedicated training games but are provided no demonstrations, documentation, or rewards. Agents must ground themselves in the environment through self-directed exploration and by inferring actionable knowledge from their own experience. At test time, agents must complete short-horizon tasks that evaluate their ability to navigate, interact, and engage with game-specific mechanics in unseen games. Out-of-the-box frontier models complete fewer than 50% of the 500 tasks due to failures in multimodal grounding, establishing that self-improvement methods have room to push performance. We demonstrate that contemporary approaches to self-improvement are lacking, with world modelling and autonomous skill discovery failing, and a novel strategy that uses curiosity-based exploration to write guides achieving only partial success. GameBoyWorlds-Playthrough tests end-to-end game completion in two fan-made Pokémon games. We show that while frontier models have been pre-exposed to official releases such as Pokémon Red, they lack essential information on the games in our testbed. Instead of relying on their parametric knowledge to succeed, agents must learn from their own experience and autonomously improve over the course of the playthrough. We show that a sophisticated agentic pipeline with multimodal memory and hierarchical subgoals fails to reach even the first major milestone in both games, establishing GameBoyWorlds as an ambitious target for self-improving agents.

cs.AI↗

arXiv2026

Auditing Multi-Agent LLM Reasoning Trees Outperforms Majority Vote and LLM-as-Judge

Multi-agent systems (MAS) can substantially extend the reasoning capacity of large language models (LLMs). Most MAS frameworks aggregate agent outputs via simple majority voting, discarding the evidential structure of reasoning traces. Majority voting is brittle under confabulation consensus, where agents share correlated biases and converge on the same incorrect rationale. We introduce AgentAuditor, which moves beyond frequency-based aggregation by organizing agent traces into a Reasoning Tree that explicitly represents agreements and divergences in their reasoning. AgentAuditor resolves conflicts by comparing branch-level evidence at critical divergence points, turning global adjudication into efficient, localized verification. We further propose Anti-Consensus Preference Optimization (ACPO), which trains the adjudicator with evidence-verified preference supervision to reduce conformity to misleading majority cues. Across four MAS frameworks and multiple reasoning benchmarks, AgentAuditor consistently improves aggregation performance over majority voting, with gains of up to 5% absolute accuracy while remaining token-efficient.

cs.AI↗

arXiv2026

AcCoRD: Evaluating User-Agent Collaboration Under Realistic User Preference Dynamics

User preferences in user-agent collaboration are rarely static and fully-specified upfront: preferences are formed, revealed, adjusted, and relaxed during interaction. Existing benchmarks for evaluating user-agent collaboration focus almost exclusively on resolving underspecified preferences, thereby failing to capture the richer dynamics of real-world interaction. We introduce AcCoRD, a user-agent collaboration benchmark requiring agents to handle diverse user preference dynamics in two domains: online shopping and travel planning. We evaluate five frontier LLMs under two prompting strategies: vanilla ReAct, and an uncertainty-guided variant that prompts models to identify and resolve ambiguity about user preferences. Our results reveal that frontier models can handle underspecification but struggle to satisfy preferences that emerge or evolve mid-interaction and require more sophisticated uncertainty modeling. Further, prompting alone fails to elicit the required uncertainty recognition. We release AcCoRD as a resource for developing agents that can navigate the full complexity of real-world user preferences.

cs.AI↗

arXiv2026

Vulnerable Code Search: Transferable Attack for Code Language Models

Reliable code retrieval is crucial for developer productivity and effective code reuse. However, current neural code language models (CLMs) powering search tools are susceptible to adversarial attacks targeting non-functional textual elements. In this paper, we introduce a programming language-agnostic, transferable, adversarial attack that exploits this CLM vulnerability. Our approach perturbs identifiers within a code snippet without altering the snippet's functionality to artificially align the code with a target query. We demonstrate that our attack, even when computed using smaller code embedding models, such as CodeT5+, is highly effective and transferable to larger, closed-source embedding models, like Voyage-code-3, or LLMs like Gemini-3.1-Pro. Our attack can increase the similarity between the query and arbitrary, irrelevant code snippets, consequently degrading key retrieval metrics such as the Mean Reciprocal Rank (MRR) of state-of-the-art models by up to 77%. The experimental results highlight the fragility of current code search methods and underscore the need for more robust, semantic-aware approaches.

cs.SE↗

arXiv2026

Should We Type or Talk to LLM Agents? A Comprehensive Study of Voice and Keyboard Input Perturbations

Human input reaches language models by typing or speaking, and each channel leaves a distinct signature: orthographic noise for keyboards; for voice, disfluency from conventional transcription and restructuring from AI-backed dictation tools. How do they impact an LLM's performance? In this paper we present HIVE (Human Input-Variation Engine), a suite of voice transcription perturbations and QWERTY keyboard perturbations. We use HIVE to evaluate how robust models are to these perturbations. We present seven findings. (i) Voice transcription perturbations lower accuracy across every instruction-tuned model we test, and it is the structure of the transcription rather than its fillers that carries the cost. (ii) QWERTY keyboard perturbations cost less, and a model absorbs a lot of them before accuracy falls away. (iii) Both trace back to one cause, how many of the question's tokens survive the perturbation: destroying a token is what hurts, while adding new ones alongside it costs little. (iv) The gap between the two channels appears only where the answer must be constructed or deduced; on multiple choice there is none. (v) The harm does not solely come from test-set contamination. (vi) It cannot be trained away with lightweight adaptation. (vii) A thinking budget recovers the keyboard channel almost entirely but leaves the spoken registers untouched, and compressed speech is worse with it.

cs.AI↗

arXiv2026

SAFECAST: Robust Failure Detection for VLA Policies with Contrast-Set Training and Calibration

Vision-language-action policies often fail under deployment-time distribution shifts such as clutter, distractor objects, lighting changes, novel objects, altered initial states, and reworded instructions. Hidden-state-based risk probes combined with functional conformal prediction can detect rollout failures, but their reliability depends on calibration data matching deployment conditions. We introduce SAFECAST, which leverages contrast set perturbations to improve hidden-state probe training and calibration for deployment time shift. SAFECAST statistically significantly improves failure detection ROC-AUC scores over a state of the art baseline in both real-world DROID and LIBERO simulation experiments across multiple VLM backbones. We further find that SAFECAST benefits most when both visual and language contrast set perturbations are used to augment data, and that with contrast set perturbations, sim-to-real calibration leads to better probes than using real rollout data only.

cs.RO↗

arXiv2026

DiPS: Dialogue Policy Selection for High-Stakes Persuasion Agents

Large Language Models (LLMs) often struggle with persuasion in high-stakes scenarios. People's individual personalities and concerns require tailored strategies rather than a one-size-fits-all approach. To address this challenge, we focus on a fire-rescue scenario in which an operator must persuade a resident to evacuate as a high-stakes persuasion domain and propose Dialogue Policy Selection (DiPS), a Q-learning framework to dynamically select persuasion strategies adapted to the evolving conversational context. Specifically, we train a critic, trained to maximize the chance of evacuation success, to select a persuasion policy at each turn based on the resident's recent utterances. We then evaluate DiPS against multiple baselines in both simulated and real human interactions. We find that DiPS achieves higher evacuation success than a zero-shot LLM and generic RAG-augmented approach.

cs.CL↗

arXiv2026

Bridging Language and Action: A Survey of Language-Conditioned Robot Manipulation

Language-conditioned robot manipulation is an emerging field aimed at enabling seamless communication and cooperation between humans and robotic agents by teaching robots to comprehend and execute instructions conveyed in natural language. This interdisciplinary area integrates scene understanding, language processing, and policy learning to bridge the gap between human instructions and robot actions. In this comprehensive survey, we systematically explore recent advancements in language-conditioned robot manipulation. We categorize existing methods based on the primary ways language is integrated into the robot system, namely language for state evaluation, language as a policy condition, language for cognitive planning and reasoning, and language in unified vision-language-action models. Specifically, we further analyze state-of-the-art techniques from five axes of action granularity, data and supervision regimes, system cost and latency, environments and evaluations, and task specification. Additionally, we highlight the key debates in the field. Finally, we discuss open challenges and future research directions, focusing on potentially enhancing generalization capabilities and addressing safety issues in language-conditioned robot manipulators.

cs.RO↗

arXiv2026

RaMem: Contextual Reinstatement for Long-term Agentic Memory

Long-term memory has become increasingly important for LLM agents that operate across extended interactions and evolving task contexts. Recent memory systems have made past experiences more persistent, compact, and retrievable, but retrieval alone does not ensure that a memory provides valid evidence for the current query. When experiences are compressed into reusable fragments, memories from different situations may appear equally relevant if they involve recurring entities or user states. We refer to this failure as context collapse: memories lose the surrounding context needed to judge whether they provide valid evidence for the current query. To address this problem, we propose Contextual Reinstatement for Agentic Memory (RaMem), a framework that turns retrieved memory fragments into contextually verifiable evidence. RaMem operates through four coordinated stages: (i) evidence anchoring grounds each memory in its original episodic conditions, especially event time, mention time, session span, and participants; (ii) recall condition induction derives the evidence conditions implied by the query; (iii) validity-aware retrieval uses these conditions to prioritize context-compatible memories while retaining content-relevant candidates as fallback evidence; and (iv) context-preserved synthesis keeps the selected memories' structured context available to the generator. Experiments on long-term memory benchmarks show that RaMem consistently improves performance over strong memory baselines, with average F1 gains of more than 10% across several backbones.

cs.AI↗

arXiv2026

SHRED: Retain-Set-Free Unlearning via Self-Distillation with Logit Demotion

Machine unlearning for large language models (LLMs) aims to selectively remove memorized content such as private data, copyrighted text, or hazardous knowledge, without costly full retraining. Most existing methods require a retain set of curated examples to prevent catastrophic degradation of general model utility, creating an extra data dependency that complicates deployment. We propose SHRED (Self-distillation via High-surprisal-only Retain-set-free Entropy Demotion), a retain-set-free unlearning method built on a key insight: not all tokens within a forget set instance carry memorized information equally. High-information tokens concentrate the model's memorized knowledge, while low-information tokens reflect general language competence. SHRED operates in two stages. (1) Selection: We perform a forward pass on a forget set instance, collect per-token autoregressive probabilities, and select the bottom (lowest probability, highest Shannon information) as forget positions; the remaining positions are retained as benign anchors. (2) Training: We construct modified KL targets that demote the memorized token's logit at forget positions while preserving the original distribution at benign positions. The model is then trained via a single top KL self-distillation objective that simultaneously drives forgetting and utility preservation. We evaluate SHRED across four standard unlearning benchmarks and demonstrate that it establishes a new Pareto-optimal trade-off between forget efficacy and model utility, outperforming retain-set-dependent methods. Our analysis shows that SHRED is robust against relearning attacks and membership-inference attacks, and it maintains stable utility even after many sequential unlearning runs.

cs.LG↗