2 citations · 7 across the 28 of their papers we have counts for
15 papers · 1 filter
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'…
DeLeaker: Dynamic Inference-Time Reweighting For Semantic Leakage Mitigation in Text-to-Image Models
Mor Ventura, Michael Toker, Or Patashnik +2
Text-to-Image (T2I) models have advanced rapidly, yet they remain vulnerable to semantic leakage, the unintended transfer of semantically related features between distinct entities…
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…
SAEs Are Good for Steering -- If You Select the Right Features
Dana Arad, Aaron Mueller, Yonatan Belinkov
Sparse Autoencoders (SAEs) have been proposed as an unsupervised approach to learn a decomposition of a model's latent space. This enables useful applications such as steering - in…
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…
MIB: A Mechanistic Interpretability Benchmark
Aaron Mueller, Atticus Geiger, Sarah Wiegreffe +20
How can we know whether new mechanistic interpretability methods achieve real improvements? In pursuit of lasting evaluation standards, we propose MIB, a Mechanistic Interpretabili…
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.…