works on

From the 1 of 11 linked papers with an AI index.

collaborators

11 papers

cs.LG2026

Sparse Autoencoders Encode Both Concepts and Functions: The Downstream Geometry of Feature Effects

Phu Gia Hoang, Anwoy Chatterjee, Tanmoy Chakraborty +2

The wide-scale use of sparse autoencoders (SAEs) as interpretability tools is limited by inconsistent links between SAE features and model behavior. Features with clear activation…

cs.CL2026

Hierarchical Latent Structures in Data Generation Process Unify Mechanistic Phenomena across Scale

Jonas Rohweder, Subhabrata Dutta, Iryna Gurevych

The paper argues that phenomena such as induction heads, function vectors, and the Hydra effect in Transformer language models can be explained by hierarchical latent structures in…

cs.AI2026

The FIL Hypothesis: Inductive Biases Help with Kernel Engineering

Nikolai Rozanov, Subhabrata Dutta, Preslav Nakov +1

The Bitter Lesson, which posits that general-purpose methods that scale with computation and data ultimately outperform those with built-in human knowledge, has become a dominant p…

cs.CL2026

Patches of Nonlinearity: Instruction Vectors in Large Language Models

Irina Bigoulaeva, Jonas Rohweder, Subhabrata Dutta +1

Despite the recent success of instruction-tuned language models and their ubiquitous usage, very little is known of how models process instructions internally. In this work, we add…

cs.CL2026

Expert Preference-based Evaluation of Automated Related Work Generation

Furkan Şahinuç, Subhabrata Dutta, Iryna Gurevych

Expert domain writing, such as scientific writing, typically demands extensive domain knowledge. Although large language models (LLMs) show promising potential in this task, evalua…

cs.CL2026

Many Circuits, One Mechanism: Input Variation and Evaluation Granularity in Circuit Discovery

Alireza Bayat Makou, Jingcheng Niu, Subhabrata Dutta +1

Circuit discovery methods identify subgraphs that explain specific model behaviors, and structural differences between discovered circuits are commonly interpreted as evidence of d…