arXiv Machine Learning

MicroLens: Read, Trace, and Rewire the Residual Stream with Two Lenses

MicroLens introduces an interactive viewer that overlays two decoding lenses—logit and Jacobian—onto the residual stream of a language model, allowing users to read, trace, and edit token representations on a layers‑by‑tokens grid. Experiments on Gemma‑3‑4B‑it show that the logit lens directs token outputs while the Jacobian lens influences how later tokens interpret information, enabling targeted edits. The tool reveals that key information (e.g., state or country names) can be read and manipulated early in the stream, and that correct answers often appear in the residual stream before reasoning is generated, with the Jacobian lens detecting them slightly earlier than the logit lens.

arXiv Machine Learning
Oct 16

Lexical Reference Frames: Interpreting LM Residual Streams in Native Semantics

The paper introduces Lexical Reference Frames (LexFrame) as a method to interpret language model residual streams by treating each hidden state as having a native reference frame induced by subsequent layers. Unlike fixed lenses or point‑cloud approaches, LexFrame reads a state through the model’s remaining layers using a set number of attention rounds, achieving high coherence (0.97) and accurate state reconstruction (within 0.16 nats). The approach also provides faithful attribution, token‑level steering, and potential early‑exit savings, making every hidden state readable, attributable, and steerable.

By Jian Gu, Hongyu Zhang, Chunyang Chen, Aldeida Aleti
arXiv AI
Jul 15

Visual Access Boundaries in Vision-Language Model Reasoning

arXiv:2607. 12815v1 Announce Type: new Abstract: Chain-of-Thought (CoT) prompting is widely used as a test-time scaling strategy for Vision-Language Models (VLMs), but it remains unclear what is extended when VLMs generate longer reasoning traces.

By Hiroto Osaka, Shohei Taniguchi, Gouki Minegishi, Kai Yamashita, Masahiro Suzuki, Yutaka Matsuo
arXiv AI
Oct 28

Beyond the Readout: A Latent Numerical Frame Shared Across Modality and Operation

The study investigates whether a multimodal model like Gemma4-12B-it uses a shared internal concept of numbers across different modalities and operations. By applying causal interventions to the model’s residual stream, the authors identify a small subspace where swapping coordinates changes the model’s arithmetic answer. They find that while the subspaces for text and image are not identical, a single rotation and rescaling aligns them, revealing a shared geometric structure that encodes numbers in hidden directions used to generate multi‑digit outputs over time.

By Oriol Juan Sabater, Alex Ferrando de las Morenas, Pau Rodr\'iguez, Xavier Suau
arXiv AI
Sep 30

Locating Answer-Correctness Signals in Frozen Large Language Models

The paper investigates where and how large language models encode signals that indicate answer correctness. By examining hidden states, token probabilities, residual-stream features, attention, and their combinations, the authors find that correctness signals are concentrated in the answer span and that different signal families complement each other. Fusing these signals improves robustness, especially under distribution shifts, and can be used to control retrieval in downstream tasks.

By Yuansen Liu, Yixuan Tang, Anthony Kum Hoe Tung
Hugging Face Trending Papers
Jul 14

Visual Access Boundaries in Vision-Language Model Reasoning

Chain-of-Thought (CoT) prompting is widely used as a test-time scaling strategy for Vision-Language Models (VLMs), but it remains unclear what is extended when VLMs generate longer reasoning traces. We ask whether CoT requires continued access to image tokens, or whether it mainly operates over visual information already made available earlier in the forward pass.