Imagen de Portada

The Hawthorne Effect in Reasoning Models: Evaluating and Steering Test Awareness

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Abdelnabi, Sahar, Salem, Ahmed
Formato:	Preprint
Publicado:	2025
Materias:	Computation and Language Computers and Society
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

ConVerse: Benchmarking Contextual Safety in Agent-to-Agent Conversations
por: Gomaa, Amr, et al.
Publicado: (2025)

AI Agents May Always Fall for Prompt Injections
por: Abdelnabi, Sahar, et al.
Publicado: (2026)

Get my drift? Catching LLM Task Drift with Activation Deltas
por: Abdelnabi, Sahar, et al.
Publicado: (2024)

Cooperation, Competition, and Maliciousness: LLM-Stakeholders Interactive Negotiation
por: Abdelnabi, Sahar, et al.
Publicado: (2023)

Evaluation Awareness in Language Models Has Limited Effect on Behaviour
por: Knecht, Amelie, et al.
Publicado: (2026)

QSTN: A Modular Framework for Robust Questionnaire Inference with Large Language Models
por: Kreutner, Maximilian, et al.
Publicado: (2025)

Models That Know How Evaluations Are Designed Score Safer
por: Deckenbach, Katharina, et al.
Publicado: (2026)

Evaluating Proactive Risk Awareness of Large Language Models
por: Luo, Xuan, et al.
Publicado: (2026)

Modeling Motivated Reasoning in Law: Evaluating Strategic Role Conditioning in LLM Summarization
por: Cho, Eunjung, et al.
Publicado: (2025)

Decomposing and Measuring Evaluation Awareness
por: Li, Changling, et al.
Publicado: (2026)

MalAlgoQA: Pedagogical Evaluation of Counterfactual Reasoning in Large Language Models and Implications for AI in Education
por: Liu, Naiming, et al.
Publicado: (2024)

Toward Socially Aware Vision-Language Models: Evaluating Cultural Competence Through Multimodal Story Generation
por: Mukherjee, Arka, et al.
Publicado: (2025)

XCR-Bench: A Multi-Task Benchmark for Evaluating Cultural Reasoning in LLMs
por: Kabir, Mohsinul, et al.
Publicado: (2026)

PRBench: Large-Scale Expert Rubrics for Evaluating High-Stakes Professional Reasoning
por: Akyürek, Afra Feyza, et al.
Publicado: (2025)

Are LLMs Court-Ready? Evaluating Frontier Models on Indian Legal Reasoning
por: Juvekar, Kush, et al.
Publicado: (2025)

Rethinking Test-Time Scaling for Medical AI: Model and Task-Aware Strategies for LLMs and VLMs
por: Oh, Gyutaek, et al.
Publicado: (2025)

DaKultur: Evaluating the Cultural Awareness of Language Models for Danish with Native Speakers
por: Müller-Eberstein, Max, et al.
Publicado: (2025)

DSO: Direct Steering Optimization for Bias Mitigation
por: Paes, Lucas Monteiro, et al.
Publicado: (2025)

White-Box Sensitivity Auditing with Steering Vectors
por: Cyberey, Hannah, et al.
Publicado: (2026)

Representation Surgery: Theory and Practice of Affine Steering
por: Singh, Shashwat, et al.
Publicado: (2024)

Assessing Judging Bias in Large Reasoning Models: An Empirical Study
por: Wang, Qian, et al.
Publicado: (2025)

MEDEQUALQA: Evaluating Biases in LLMs with Counterfactual Reasoning
por: Ghosh, Rajarshi, et al.
Publicado: (2025)

Steering at the Source: Style Modulation Heads for Robust Persona Control
por: Izawa, Yoshihiro, et al.
Publicado: (2026)

JudgeMeNot: Personalizing Large Language Models to Emulate Judicial Reasoning in Hebrew
por: Razumenko, Itay, et al.
Publicado: (2026)

Evaluating Cultural Awareness of LLMs for Yoruba, Malayalam, and English
por: Dawson, Fiifi, et al.
Publicado: (2024)

A Theory of Response Sampling in LLMs: Part Descriptive and Part Prescriptive
por: Sivaprasad, Sarath, et al.
Publicado: (2024)

Translate With Care: Addressing Gender Bias, Neutrality, and Reasoning in Large Language Model Translations
por: Zahraei, Pardis Sadat, et al.
Publicado: (2025)

Steering the CensorShip: Uncovering Representation Vectors for LLM "Thought" Control
por: Cyberey, Hannah, et al.
Publicado: (2025)

Inference-Time Reasoning Selectively Reduces Implicit Social Bias in Large Language Models
por: Apsel, Molly, et al.
Publicado: (2026)

In-Situ Behavioral Evaluation for LLM Fairness, Not Standardized-Test Scores
por: Tang, Zeyu, et al.
Publicado: (2026)

The Reasoning Trap -- Logical Reasoning as a Mechanistic Pathway to Situational Awareness
por: Sahoo, Subramanyam, et al.
Publicado: (2026)

Taxonomy, Opportunities, and Challenges of Representation Engineering for Large Language Models
por: Wehner, Jan, et al.
Publicado: (2025)

Reasoning Language Models for complex assessments tasks: Evaluating parental cooperation from child protection case reports
por: Stoll, Dragan, et al.
Publicado: (2026)

Extrinsic Evaluation of Cultural Competence in Large Language Models
por: Bhatt, Shaily, et al.
Publicado: (2024)

Out of the Box, into the Clinic? Evaluating State-of-the-Art ASR for Clinical Applications for Older Adults
por: van Dijk, Bram, et al.
Publicado: (2025)

Agent-to-Agent Theory of Mind: Testing Interlocutor Awareness among Large Language Models
por: Choi, Younwoo, et al.
Publicado: (2025)

Narrative over Numbers: The Identifiable Victim Effect and its Amplification Under Alignment and Reasoning in Large Language Models
por: Raiyan, Syed Rifat
Publicado: (2026)

Firewalls to Secure Dynamic LLM Agentic Networks
por: Abdelnabi, Sahar, et al.
Publicado: (2025)

Evaluating Implicit Biases in LLM Reasoning through Logic Grid Puzzles
por: Jahara, Fatima, et al.
Publicado: (2025)

RDBE: Reasoning Distillation-Based Evaluation Enhances Automatic Essay Scoring
por: Mohammadkhani, Ali Ghiasvand
Publicado: (2024)