Home Knowledge Base SAE International's

SAE International's

No mentions found

This entity hasn't been tracked yet, or Iris is still building its knowledge base.

Related Articles from SNS

Sparse Autoencoders Reveal Interpretable and Steerable Features in VLA Models

Announce Type: replace Abstract: Vision-Language-Action (VLA) models have emerged as a promising approach for general-purpose robot manipulation. However, little research has mechanistically explored when and why they generalize across objects, scenes, and instructions. To probe internal representations, we train Sparse Autoencoders (SAEs) on the VLA's hidden-layer activations.

arXiv CS 1d ago

The Tell-Tale Norm: $\ell_2$ Magnitude as a Signal for Reasoning Dynamics in Large Language Models

arXiv:2606.06188v1 Announce Type: new Abstract: Recent work has sought to understand Large Language Models (LLMs) reasoning, yet a principled, model-intrinsic signal that captures its layer-wise reasoning dynamics remains underexplored. We bridge this gap by demonstrating that the l2 norm of hidden states serves as an endogenous signal of the model's reasoning intensity. Using Sparse Autoencoders (SAEs) as a diagnostic probe, we observe that LLMs' internal reasoning is marked by a sharp...

arXiv CS 5d ago

Steering LLMs? Actually, Sparse Autoencoders can outperform simple baselines

arXiv:2605.31183v1 Announce Type: new Abstract: Sparse Autoencoders (SAEs) have been seen as a promising avenue for exploring the internals of Large Language Models (LLMs) and for steering model output generation. When AxBench - a model steering benchmark - was introduced in Wu et al. (2025), SAEs did not seem to live up to their original hype due to poor steering performance relative to a set of simple baselines.

arXiv CS 9d ago

Slate Auto gets serious about privacy for its bare-bones EV pickup

Slate Auto may be one of the most interesting companies in the American automotive industry right now. Based in Warsaw, Indiana, the startup is taking a completely different approach to building an electric pickup truck. Forget Ford's clean-sheet "skunk works" story; the Slate Truck's design has been stripped down to just 600 parts and components.

Ars Technica 8d ago

Beyond the Black Box: Interpretability of Agentic AI Tool Use

Announce Type: replace Abstract: AI agents are promising for high-stakes enterprise workflows, but dependable deployment remains limited because tool-use failures are difficult to diagnose and control. Agents may skip required tool calls, invoke tools unnecessarily, or take actions whose consequence becomes visible only after execution. Existing observability methods are external: prompts reveal correlations, evaluations score outputs, and logs arrive only after the model has already acted.

arXiv CS 2d ago

When Models Refuse: Political Steerability and Feature Richness as Measures of Ideological Depth

Announce Type: replace Abstract: Large language models (LLMs) sometimes refuse to follow benign instructions, such as declining to argue a political position or adopt a stated persona, and such refusals are commonly read as safety guardrails at work. We ask whether they can instead signal a **capability deficit**: a shortage of the internal representations a model needs to reason from the instructed perspective. To investigate, we introduce **ideological depth**, a property with two...

arXiv CS 7d ago

AdaptiveK: Complexity-Driven Sparse Autoencoders for Interpretable Language Model Representations

arXiv:2508.17320v3 Announce Type: replace Abstract: Understanding the internal representations of large language models (LLMs) remains a central challenge for interpretability research. Sparse autoencoders (SAEs) offer a promising solution by decomposing activations into interpretable features, but existing approaches rely on fixed sparsity constraints that fail to account for input complexity. We propose AdaptiveK SAE (Adaptive Top K Sparse Autoencoders), a novel framework that dynamically...

arXiv CS 8d ago

Whisper Hallucination Detection and Mitigation via Hidden Representation Steering and Sparse AutoEncoders

arXiv:2606.07473v1 Announce Type: new Abstract: Whisper, a widely adopted ASR model, is known to suffer from hallucinations - coherent transcriptions generated for non-speech audio entirely disconnected from the input. We investigate whether hallucinations can be detected and mitigated through Whisper's internal representations. We extract audio encoder activations and evaluate two representation spaces: raw Whisper activations and Sparse AutoEncoder (SAE) latents.

arXiv CS 2d ago

Sparse Autoencoders for Interpretable Emotion Control in Text-to-Speech

arXiv:2606.01479v1 Announce Type: new Abstract: Integrating large language models (LLMs) into text-to-speech (TTS) systems has improved speech expressiveness, yet interpretable emotional control remains challenging. Existing approaches primarily rely on external conditioning or global activation steering, offering limited insight into the internal representations underlying emotional control.

arXiv CS 8d ago

Doctors thought this kidney drug helped some patients. It may help millions more.

Doctors thought this kidney drug helped some patients. It may help millions more. - Date: - June 8, 2026

Science Daily 2d ago