Markus Knauer, Samuel Bustamante, Thomas Eiband, Alin Albu-Schäffer, Freek Stulp, João Silvério · Mar 4, 2026 · Citations: 0
DemonstrationsGeneral
We demonstrate the framework on a 7-DoF torque-controlled robot performing an industrial bearing ring insertion task, showing successful skill adaptation through natural language commands for speed adjustment, trajectory correction, and…
Angana Borah, Zohaib Khan, Rada Mihalcea, Verónica Pérez-Rosas · Mar 3, 2026 · Citations: 0
Automatic MetricsSimulation EnvGeneral
As Large Language Models (LLMs) are increasingly used to simulate human behaviors, we investigate whether they can simulate demographic misinformation susceptibility, treating beliefs as a primary driving factor.
We study prompt-based conditioning and post-training adaptation, and conduct a multi-fold evaluation using: (i) susceptibility accuracy and (ii) counterfactual demographic sensitivity.
Alejandro Breen Herrera, Aayush Sheth, Steven G. Xu, Zhucheng Zhan, Charles Wright, Marcus Yearwood · Mar 3, 2026 · Citations: 0
Pairwise PreferenceRubric RatingLlm As JudgeSimulation EnvGeneral
Conversational shopping assistants (CSAs) represent a compelling application of agentic AI, but moving from prototype to production reveals two underexplored challenges: how to evaluate multi-turn interactions and how to optimize tightly…
We introduce a multi-faceted evaluation rubric that decomposes end-to-end shopping quality into structured dimensions and develop a calibrated LLM-as-judge pipeline aligned with human annotations.
Ashutosh Chaubey, Jiacheng Pang, Mohammad Soleymani · Mar 3, 2026 · Citations: 0
Pairwise PreferenceAutomatic MetricsGeneral
In this work, we propose Modality-Decoupled Direct Preference Optimization (MoD-DPO), a simple and effective framework for improving modality grounding in omni LLMs.
Extensive experiments across multiple audiovisual hallucination benchmarks demonstrate that MoD-DPO consistently improves perception accuracy and hallucination resistance, outperforming previous preference optimization baselines under…
Zhi Xu, Jiaqi Li, Xiaotong Zhang, Hong Yu, Han Liu · Mar 3, 2026 · Citations: 0
Red TeamGeneral
Large language models (LLMs) have achieved remarkable success across diverse applications but remain vulnerable to jailbreak attacks, where attackers craft prompts that bypass safety alignment and elicit unsafe responses.
Eliseo Bao, Anxo Perez, Xi Wang, Javier Parapar · Mar 3, 2026 · Citations: 0
Llm As JudgeSimulation EnvGeneral
Large Language Model (LLM) personas with explicit specifications of attributes, background, and behavioural tendencies are increasingly used to simulate human conversations for tasks such as user modeling, social reasoning, and behavioural…
Ensuring that persona-grounded simulations faithfully reflect human conversational behaviour is therefore critical.
Zhongxi Wang, Yueqian Lin, Jingyang Zhang, Zinuo Cheng, Hai Helen Li, Yiran Chen · Mar 3, 2026 · Citations: 0
Automatic MetricsGeneral
We present MUSE (Multimodal Unified Safety Evaluation), an open-source, browser-based, run-centric platform for multimodal safety evaluation.
Across 11,700 evaluations on six multimodal LLMs, direct text-only requests yield only 3.1% macro hard ASR and 4.4% soft ASR, while iterative attack procedures are substantially more effective.
The rapid proliferation of Claude agent skills has raised the central question of how to effectively leverage, manage, and scale the agent skill ecosystem.
In this paper, we propose AgentSkillOS, the first principled framework for skill selection, orchestration, and ecosystem-level management.
Chenxiao Yang, Nathan Srebro, Zhiyuan Li · Mar 2, 2026 · Citations: 0
Automatic MetricsGeneral
We further generalize our framework to modern agentic systems with arbitrary context processing and control flows, and prove that recursive models can achieve optimal power within this broader class.
Guilhem Fouilhé, Rebecca Eifler, Antonin Poché, Sylvie Thiébaux, Nicholas Asher · Mar 2, 2026 · Citations: 0
Pairwise PreferenceGeneral
When automating plan generation for a real-world sequential decision problem, the goal is often not to replace the human planner, but to facilitate an iterative reasoning and elicitation process, where the human's role is to guide the AI…
To enable natural interaction with such a system, we present a multi-agent Large Language Model (LLM) architecture that is agnostic to the explanation framework and enables user- and context-dependent interactive explanations.
Cheng Jiayang, Dongyu Ru, Lin Qiu, Yiyang Li, Xuezhi Cao, Yangqiu Song · Mar 2, 2026 · Citations: 0
Simulation EnvGeneral
Long-horizon interactions between users and LLM-based assistants necessitate effective memory management, yet current approaches face challenges in training and evaluation of memory.
To address these gaps, we introduce AMemGym, an interactive environment enabling on-policy evaluation and optimization for memory-driven personalization.
Extensive experiments on both in-domain and out-of-domain reasoning-intensive benchmarks demonstrate that LaSER significantly outperforms state-of-the-art baselines.
DARS incorporates a multi-stage rhythm predictor optimized by contrastive preferences between normal and dysarthric speech, along with a dysarthric-style conditional flow matching mechanism, jointly enhancing temporal rhythm reconstruction…
Achuth Chandrasekhar, Janghoon Ock, Amir Barati Farimani · Mar 1, 2026 · Citations: 0
Automatic MetricsGeneral
In this work, we introduce Catalyst-Agent, a Model Context Protocol (MCP) server-based, LLM-powered AI agent.
Catalyst-Agent achieves a success rate of 33-41% among all the materials it chooses and evaluates, and manages to converge in 1-4 trials per successful material on average.
Shiqi Chen, Jingze Gai, Ruochen Zhou, Jinghan Zhang, Tongyao Zhu, Junlong Li · Feb 28, 2026 · Citations: 0
Automatic MetricsGeneral
Real-world tool-using agents operate over long-horizon workflows with recurring structure and diverse demands, where effective behavior requires not only invoking atomic tools but also abstracting, and reusing higher-level tool…
Evaluating state-of-the-art agents on SkillCraft, we observe substantial efficiency gains, with token usage reduced by up to 80% by skill saving and reuse.
Alokesh Manna, William Snyder, Whitney Tabor · Feb 27, 2026 · Citations: 0
Pairwise PreferenceGeneral
We trained Meta's OPT model on the 100M word BabyLM dataset, and evaluated it on the BLiMP benchmark, which consists of 67 classes, each defined by sentence pairs that differ in a targeted syntactic or semantic rule violation.
We tested the model's preference for grammatical over ungrammatical sentences across training iterations and grammatical types.
