2 citations · 7 across the 28 of their papers we have counts for
22 papers · 1 filter
Seeing the Forest for the Trees: Emergent Agreement from Individual Conversations
Adi Simhi, Yaniv Nikankin, Jonathan Herzig +3
As LLM agents interact in the wild and shape each other's decisions, understanding how they converge toward agreement becomes important. We study this in multi-turn conversations w…
Directing large language models to follow the letter or spirit of the law
Peng Qian, Andrew Li, Sam Chen +3
The distinction between the spirit and letter of the law is a central issue across research and everyday life, and a growing concern for building safe, intelligent machines. What i…
Do LLMs Choose Like Humans? Using Cognitive Theory to Evaluate LLM Decision-Making
Johnathan Sun, Andrei Shleifer, Yonatan Belinkov
Large language models (LLMs) exhibit a range of human-like decision-making behaviors, but whether these reflect similar underlying mechanisms or surface-level mimicry remains uncle…
Differentiable Faithfulness Alignment for Cross-Model Circuit Transfer
Shun Shao, Binxu Wang, Shay B. Cohen +2
Mechanistic interpretability has made it possible to localize circuits underlying specific behaviors in language models, but existing methods are expensive, model-specific, and dif…
Masked by Consensus: Disentangling Privileged Knowledge in LLM Correctness
Tomer Ashuach, Shai Gretz, Yoav Katz +2
Humans use introspection to evaluate their understanding through private internal states inaccessible to external observers. We investigate whether large language models possess si…
Reasoning Models Know What's Important, and Encode It in Their Activations
Yaniv Nikankin, Martin Tutek, Tomer Ashuach +2
Language models often solve complex tasks by generating long reasoning chains, consisting of many steps with varying importance. While some steps are crucial for generating the fin…
From Feelings to Metrics: Understanding and Formalizing How Users Vibe-Test LLMs
Itay Itzhak, Eliya Habba, Gabriel Stanovsky +1
Evaluating LLMs is challenging, as benchmark scores often fail to capture models' real-world usefulness. Instead, users often rely on ``vibe-testing'': informal experience-based ev…
Large Language Models Generate Harmful Responses Using a Distinct Mechanism, Shared Across Harm Types
Hadas Orgad, Boyi Wei, Kaden Zheng +4
Large language models remain vulnerable to jailbreaks that elicit harmful responses, yet the mechanism behind harmful response generation is poorly understood. Here, we investigate…
Old Habits Die Hard: How Conversational History Geometrically Traps LLMs
Adi Simhi, Fazl Barez, Martin Tutek +2
How does the conversational past of large language models (LLMs) influence their future performance? Recent work suggests that LLMs are affected by their conversational history in…
Findings of the BlackboxNLP 2025 Shared Task: Localizing Circuits and Causal Variables in Language Models
Dana Arad, Yonatan Belinkov, Hanjie Chen +5
Mechanistic interpretability (MI) seeks to uncover how language models (LMs) implement specific behaviors, yet measuring progress in MI remains challenging. The recently released M…
BlackboxNLP-2025 MIB Shared Task: Improving Circuit Faithfulness via Better Edge Selection
Yaniv Nikankin, Dana Arad, Itay Itzhak +4
One of the main challenges in mechanistic interpretability is circuit discovery, determining which parts of a model perform a given task. We build on the Mechanistic Interpretabili…
HACK: Hallucinations Along Certainty and Knowledge Axes
Adi Simhi, Jonathan Herzig, Itay Itzhak +7
Hallucinations in LLMs present a critical barrier to their reliable usage. Existing research usually categorizes hallucination by their external properties rather than by the LLMs'…
Silent Tokens, Loud Effects: Padding in LLMs
Rom Himelstein, Amit LeVi, Yonatan Belinkov +1
Padding tokens are widely used in large language models (LLMs) to equalize sequence lengths during batched inference. While they should be fully masked, implementation errors can c…
Global PIQA: Evaluating Commonsense Reasoning Across 100+ Languages and Cultures
Tyler A. Chang, Catherine Arnett, Abdelrahman Sadallah +377
To date, there exist almost no culturally-specific evaluation benchmarks for large language models (LLMs) that cover a large number of languages and cultures. In this paper, we pre…
ManagerBench: Evaluating the Safety-Pragmatism Trade-off in Autonomous LLMs
Adi Simhi, Jonathan Herzig, Martin Tutek +3
As large language models (LLMs) evolve from conversational assistants into autonomous agents, evaluating the safety of their actions becomes critical. Prior safety benchmarks have…
CRISP: Persistent Concept Unlearning via Sparse Autoencoders
Tomer Ashuach, Dana Arad, Aaron Mueller +2
As large language models (LLMs) are increasingly deployed in real-world applications, the need to selectively remove unwanted knowledge while preserving model utility has become pa…
Same Task, Different Circuits: Disentangling Modality-Specific Mechanisms in VLMs
Yaniv Nikankin, Dana Arad, Yossi Gandelsman +1
Vision-Language models (VLMs) show impressive abilities to answer questions on visual inputs (e.g., counting objects in an image), yet demonstrate higher accuracies when performing…
Language Models use Lookbacks to Track Beliefs
Nikhil Prakash, Natalie Shapira, Arnab Sen Sharma +5
How do language models (LMs) represent characters' beliefs, especially when those beliefs may differ from reality? This question lies at the heart of understanding the Theory of Mi…
Follow the Flow: On Information Flow Across Textual Tokens in Text-to-Image Models
Guy Kaplan, Michael Toker, Yuval Reif +2
Text-to-image generation models suffer from alignment problems, where generated images fail to accurately capture the objects and relations in the text prompt. Prior work has focus…
Are formal and functional linguistic mechanisms dissociated in language models?
Michael Hanna, Yonatan Belinkov, Sandro Pezzelle
Although large language models (LLMs) are increasingly capable, these capabilities are unevenly distributed: they excel at formal linguistic tasks, such as producing fluent, gramma…
Measuring Chain of Thought Faithfulness by Unlearning Reasoning Steps
Martin Tutek, Fateme Hashemi Chaleshtori, Ana Marasović +1
When prompted to think step-by-step, language models (LMs) produce a chain of thought (CoT), a sequence of reasoning steps that the model supposedly used to produce its prediction.…
Answer, Assemble, Ace: Understanding How LMs Answer Multiple Choice Questions
Sarah Wiegreffe, Oyvind Tafjord, Yonatan Belinkov +2
Multiple-choice question answering (MCQA) is a key competence of performant transformer language models that is tested by mainstream benchmarks. However, recent evidence shows that…