activity
20242026
collaborators

8 papers

cs.LG2026

SynthSAEBench: Evaluating Sparse Autoencoders on Scalable Realistic Synthetic Data

David Chanin, Adrià Garriga-Alonso

Improving Sparse Autoencoders (SAEs) requires benchmarks that can precisely validate architectural innovations. Current LLM-based SAE benchmarks are too noisy to differentiate arch…

cs.LG2026

Path Channels and Plan Extension Kernels: a Mechanistic Description of Planning in a Sokoban RNN

Mohammad Taufeeque, Aaron David Tucker, Adam Gleave +1

We partially reverse-engineer a convolutional recurrent neural network (RNN) trained with model-free reinforcement learning to play the box-pushing game Sokoban. We find that the R…

cs.AI2026

Post-Hoc Reasoning in Chain of Thought: Decoding and Steering Pre-Committed Answers

Kyle Cox, Darius Kianersi, Adrià Garriga-Alonso

As chain of thought (CoT) has become central to scaling reasoning capabilities in large language models (LLMs), it has also emerged as a promising tool for interpretability, sugges…

cs.AI2026

Among Us: A Sandbox for Measuring and Detecting Agentic Deception

Satvik Golechha, Adrià Garriga-Alonso

Prior studies on deception in language-based AI agents typically assess whether the agent produces a false statement about a topic, or makes a binary choice prompted by a goal, rat…

cs.LG2025

DiFR: Inference Verification Despite Nondeterminism

Adam Karvonen, Daniel Reuter, Roy Rinberg +3

As demand for LLM inference grows, it is becoming increasingly important that providers and their customers can verify that inference processes are performed correctly, without err…

cs.LG2025

Planning in a recurrent neural network that plays Sokoban

Mohammad Taufeeque, Philip Quirke, Maximilian Li +4

Planning is essential for solving complex tasks, yet the internal mechanisms underlying planning in neural networks remain poorly understood. Building on prior work, we analyze a r…