Here are the accepted posters for the Workshop on Actionable Interpretability 2026:
Session 1 (10:50 - 11:55)
29 - When the Loop Closes Quietly: Mechanistic Interpretability as Contestability Infrastructure for Deployed AI
Allan Ochola
Poster Session 1
52 - On the Non-Identifiability of Steering Vectors in Large Language Models
Sohan Venkatesh, Ashish Mahendran Kurapath
Poster Session 1
69 - Mechanistic Specialization Does Not Guarantee Performance: Why GPT-2 Outperforms Dual Attention Transformers on Identity Rule Tasks
Jonathan Scott, Leilani H. Gilpin, Adrian Brasoveanu
Poster Session 1
76 - Generative Meta-Models are Adversarial Classifiers
Diego Calanzone, Pietro Greiner, Pierre-Luc Bacon
Poster Session 1
88 - LangFIR: Discovering Sparse Language-Specific Features from Monolingual Data for Language Steering
Sing Hieng Wong, Hassan Sajjad, A.B. Siddique
Poster Session 1
97 - Targeting the Attention Heads Behind Object Hallucination in LLaVA
Armaan Sandhu, Abhilasha Senapati, Hima Jagadeesha Kammachi
Poster Session 1
109 - Guardrail Vectors: Bidirectional Steering and Framing-Induced Safety Failures
Paul Tylkin
Poster Session 1
116 - LayerRoPE: Dynamic Depth-wise Magnitude & Angular Superposition
Shikhar Srivastava, Christopher Kanan
Poster Session 1
129 - Are Latent Reasoning Models Easily Interpretable?
Connor Dilgren, Sarah Wiegreffe
Poster Session 1
Session 2 (14:25 - 15:30)
40 - Simple Activation Clipping Defeats Circuit Breakers
George Yakushev, Kirill Pupkov
Poster Session 2
50 - Sycophantic Agreement Transfers with Neutral Data via Contrastive Preference Optimization
Camila Blank, Zhuofan Ying, Christopher Potts, Peter Hase, Jing Huang
Poster Session 2
62 - Before You Act: Benchmarking Interpretability in Vision-Language Models
Dia'a AL-Dweikat, Abdullah AlDahoul, Oyedeji Sonuga, Artur Shyutts, Jiahang He, Kevin Zhu, Ruizhe Li, Anusha Mujumdar
Poster Session 2
66 - Shifting Mechanisms: How Positional Encoding Choice Shapes Long-Context Retrieval
Eric Enouen, sainyam galhotra
Poster Session 2
86 - Limits of Activation-Direction Interventions for Emergent Misalignment Beyond Rank-1 LoRA
Jayden Ramirez, Santiago Gutierrez, Anishansu Pradhan, Hikaru Isayama, Tinuade Adeleke
Poster Session 2
89 - Facial and Vocal Emotion Elicit Partially Shared Emotion-Sensitive Neurons in Multimodal Foundation Models
Xiutian Zhao, Luqi Sun, Berrak Sisman
Poster Session 2
101 - MoRFI: Monotonic Sparse Autoencoder Feature Identification
Dimitris Dimakopoulos, Shay B. Cohen, Ioannis Konstas
Poster Session 2
107 - Generalization of Steering Vectors for Chain-of-Thought Faithfulness
Matthew Nguyen, Kyle Cox, Austin Meek, Iván Arcuschin
Poster Session 2
113 - Who Is Left of Whom? Tracing Spatial Evidence and Role Binding in Relative-Position Reasoning
Yingjin Song, Denis Paperno, Albert Gatt
Poster Session 2
128 - Reasoning Fine-Tuning Induces Persistent Latent Policy States
Abir Harrasse, Michael Lan, Hunar Batra, Fateme Hashemi Chaleshtori, Chaithanya Bandi
Poster Session 2
Online only
16 - From Refusal Tokens to Refusal Control: Discovering and Steering Category-Specific Refusal Directions
Rishab Alagharu, Ishneet Sukhvinder Singh, Shaibi Shamsudeen, Zhen Wu, Ashwinee Panda
Online Only
28 - Actionable Activation Directions for Detecting and Mitigating Emergent Misalignment Across Language Model Families
Abdul Rafay Syed
Online Only
55 - Don’t Blink: Evidence Collapse during Multimodal Reasoning
Suresh Raghu, Satwik Pandey
Online Only
59 - Tuning Language Models by Mixture-of-Depths Ensemble
Haoyan Luo, Lucia Specia
Online Only
77 - CircuitSteer: Geometrically Aligned Multi-Layer Steering via Sparse Autoencoder Circuits
Mehrshad Saadatinia, Parsa Razmara, Ardalan Aryashad, Ali Abbasi, Seyedarmin Azizi
Online Only
82 - Mechanistic Analysis of Safety Disparities Across Vision and Text Modalities in Vision-Language Models
Imran Kutianawala, Sarthak Tayal, Melissa Yu, Ayda Sultan
Online Only
103 - RARE: Decoupling Representation Steering from Expert Routing in Mixture-of-Experts Language Models
Zhibo Zhang, Zhen Ouyang, Ling Shi, Kailong Wang
Online Only
