Cover Image

Tracing Multilingual Representations in LLMs with Cross-Layer Transcoders

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Harrasse, Abir, Draye, Florent, Pandey, Punya Syon, Jin, Zhijing, Schölkopf, Bernhard
Format:	Preprint
Published:	2025
Subjects:	Computation and Language
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

CLT-Forge: A Scalable Library for Cross-Layer Transcoders and Attribution Graphs
by: Draye, Florent, et al.
Published: (2026)

BinaryPPO: Efficient Policy Optimization for Binary Classification
by: Pandey, Punya Syon, et al.
Published: (2026)

Accidental Vulnerability: Factors in Fine-Tuning that Shift Model Safeguards
by: Pandey, Punya Syon, et al.
Published: (2025)

CORE: Measuring Multi-Agent LLM Interaction Quality under Game-Theoretic Pressures
by: Pandey, Punya Syon, et al.
Published: (2025)

Identifying Intervenable and Interpretable Features via Orthogonality Regularization
by: Miller, Moritz, et al.
Published: (2026)

Quriosity: Analyzing Human Questioning Behavior and Causal Inquiry through Curiosity-Driven Queries
by: Ceraolo, Roberto, et al.
Published: (2024)

SocialHarmBench: Revealing LLM Vulnerabilities to Socially Harmful Requests
by: Pandey, Punya Syon, et al.
Published: (2025)

How Robust Are Router-LLMs? Analysis of the Fragility of LLM Routing Capabilities
by: Kassem, Aly M., et al.
Published: (2025)

Preserving Historical Truth: Detecting Historical Revisionism in Large Language Models
by: Ortu, Francesco, et al.
Published: (2026)

Improving Large Language Model Safety with Contrastive Representation Learning
by: Simko, Samuel, et al.
Published: (2025)

Competition of Mechanisms: Tracing How Language Models Handle Facts and Counterfactuals
by: Ortu, Francesco, et al.
Published: (2024)

Debate, Deliberate, Decide (D3): A Cost-Aware Adversarial Framework for Reliable and Interpretable LLM Evaluation
by: Harrasse, Abir, et al.
Published: (2024)

The Odyssey of Commonsense Causality: From Foundational Benchmarks to Cutting-Edge Reasoning
by: Cui, Shaobo, et al.
Published: (2024)

Do LLMs Think Fast and Slow? A Causal Study on Sentiment Analysis
by: Lyu, Zhiheng, et al.
Published: (2024)

Objective Matters: Fine-Tuning Objectives Shape Safety, Robustness, and Persona Drift
by: Vennemeyer, Daniel, et al.
Published: (2026)

A diverse Multilingual News Headlines Dataset from around the World
by: Leeb, Felix, et al.
Published: (2024)

Are LLMs Good Safety Agents or a Propaganda Engine?
by: Yadav, Neemesh, et al.
Published: (2025)

Democratic or Authoritarian? Probing a New Dimension of Political Biases in Large Language Models
by: Piedrahita, David Guzman, et al.
Published: (2025)

DARS: Dynamic Action Re-Sampling to Enhance Coding Agent Performance by Adaptive Tree Traversal
by: Aggarwal, Vaibhav, et al.
Published: (2025)

Language Model Alignment in Multilingual Trolley Problems
by: Jin, Zhijing, et al.
Published: (2024)

Cooperate or Collapse: Emergence of Sustainable Cooperation in a Society of LLM Agents
by: Piatti, Giorgio, et al.
Published: (2024)

Analyzing the Role of Semantic Representations in the Era of Large Language Models
by: Jin, Zhijing, et al.
Published: (2024)

Exploring the Jungle of Bias: Political Bias Attribution in Language Models via Dependency Analysis
by: Jenny, David F., et al.
Published: (2023)

Are Language Models Consequentialist or Deontological Moral Reasoners?
by: Samway, Keenan, et al.
Published: (2025)

When Ethics and Payoffs Diverge: LLM Agents in Morally Charged Social Dilemmas
by: Backmann, Steffen, et al.
Published: (2025)

Corrupted by Reasoning: Reasoning Language Models Become Free-Riders in Public Goods Games
by: Piedrahita, David Guzman, et al.
Published: (2025)

When Do Language Models Endorse Limitations on Human Rights Principles?
by: Samway, Keenan, et al.
Published: (2026)

Transcoders Trace Visual Grounding and Hallucinations in Vision-Language Models
by: Damianos, Dimitrios, et al.
Published: (2026)

Prune, Interpret, Evaluate: A Cross-Layer Transcoder-Native Framework for Efficient Circuit Discovery via Feature Attribution
by: Chen, Qinhao, et al.
Published: (2026)

Causal Responsibility Attribution for Human-AI Collaboration
by: Qi, Yahang, et al.
Published: (2024)

The Curious Case of Curiosity across Human Cultures and LLMs
by: Borah, Angana, et al.
Published: (2025)

CausalCite: A Causal Formulation of Paper Citations
by: Kumar, Ishan, et al.
Published: (2023)

Can Theoretical Physics Research Benefit from Language Agents?
by: Lu, Sirui, et al.
Published: (2025)

Can Large Language Models Infer Causation from Correlation?
by: Jin, Zhijing, et al.
Published: (2023)

Implicit Personalization in Language Models: A Systematic Study
by: Jin, Zhijing, et al.
Published: (2024)

Causality for Natural Language Processing
by: Jin, Zhijing
Published: (2025)

Nullpointer at CheckThat! 2024: Identifying Subjectivity from Multilingual Text Sequence
by: Biswas, Md. Rafiul, et al.
Published: (2024)

Cross-Lingual Auto Evaluation for Assessing Multilingual LLMs
by: Doddapaneni, Sumanth, et al.
Published: (2024)

Intrinsically Interpretable Attention via Sparse Post-Training
by: Draye, Florent, et al.
Published: (2025)

LLMs Beyond English: Scaling the Multilingual Capability of LLMs with Cross-Lingual Feedback
by: Lai, Wen, et al.
Published: (2024)