English
Related papers

Related papers: Latent Introspection: Models Can Detect Prior Conc…

200 papers

Answering multi-hop reasoning questions requires retrieving and synthesizing information from diverse sources. Language models (LMs) struggle to perform such reasoning consistently. We propose an approach to pinpoint and rectify multi-hop…

Computation and Language · Computer Science 2024-11-11 Mansi Sakarvadia

Understanding which inductive biases could be helpful for the unsupervised learning of object-centric representations of natural scenes is challenging. In this paper, we systematically investigate the performance of two models on datasets…

Computer Vision and Pattern Recognition · Computer Science 2022-08-16 Samuele Papa , Ole Winther , Andrea Dittadi

We study unsupervised learning by developing introspective generative modeling (IGM) that attains a generator using progressively learned deep convolutional neural networks. The generator is itself a discriminator, capable of introspection:…

Computer Vision and Pattern Recognition · Computer Science 2017-04-26 Justin Lazarow , Long Jin , Zhuowen Tu

Large language models (LLMs) suffer from proactive interference (PI): outdated information in the context window disrupts retrieval of current values. This interference degrades retrieval accuracy log-linearly as stale associations…

Artificial Intelligence · Computer Science 2026-03-17 Ying Xie

Prior work has shown that a significant driver of performance in reasoning models is their ability to reason and self-correct. A distinctive marker in these reasoning traces is the token wait, which often signals reasoning behavior such as…

Artificial Intelligence · Computer Science 2025-10-07 Dmitrii Troitskii , Koyena Pal , Chris Wendler , Callum Stuart McDougall , Neel Nanda

Despite the remarkable generation capabilities of diffusion models, recent studies have shown that they can memorize and create harmful content when given specific text prompts. Although fine-tuning approaches have been developed to…

Computer Vision and Pattern Recognition · Computer Science 2025-06-03 Siyi Chen , Yimeng Zhang , Sijia Liu , Qing Qu

Autoregressive Sequence-To-Sequence models are the foundation of many Deep Learning achievements in major research fields such as Vision and Natural Language Processing. Despite that, they still present significant limitations. For…

Computation and Language · Computer Science 2024-08-27 Jia Cheng Hu , Roberto Cavicchioli , Alessandro Capotondi

We consider the feedback capacity of a MIMO channel whose channel output is given by a linear state-space model driven by the channel inputs and a Gaussian process. The generality of our state-space model subsumes all previous studied…

Information Theory · Computer Science 2022-07-22 Oron Sabag , Victoria Kostina , Babak Hassibi

Concept Bottleneck Models (CBMs) provide a basis for semantic abstractions within a neural network architecture. Such models have primarily been seen through the lens of interpretability so far, wherein they offer transparency by inferring…

Computer Vision and Pattern Recognition · Computer Science 2025-12-09 Deepika SN Vemuri , Gautham Bellamkonda , Aditya Pola , Vineeth N Balasubramanian

Deep learning representations are often difficult to interpret, which can hinder their deployment in sensitive applications. Concept Bottleneck Models (CBMs) have emerged as a promising approach to mitigate this issue by learning…

Machine Learning · Computer Science 2026-01-30 Antonio Almudévar , José Miguel Hernández-Lobato , Alfonso Ortega

Part-based reasoning is a classical strategy to make a computer vision model directly focus on the object parts that are relevant to the downstream task. In the context of deep learning, this also serves to improve by-design…

Computer Vision and Pattern Recognition · Computer Science 2026-05-08 Ananthu Aniraj , Cassio F. Dantas , Dino Ienco , Massimiliano Mancini , Diego Marcos

In this work, we propose an introspection technique for deep neural networks that relies on a generative model to instigate salient editing of the input image for model interpretation. Such modification provides the fundamental…

Machine Learning · Computer Science 2019-07-09 Shusen Liu , Bhavya Kailkhura , Donald Loveland , Yong Han

Large language models (LLMs) process entire input contexts indiscriminately, which is inefficient when the information required to answer a query is localized within the context. We present dynamic context cutoff, a novel method enabling…

Computation and Language · Computer Science 2026-02-10 Roy Xie , Junlin Wang , Paul Rosu , Chunyuan Deng , Bolun Sun , Zihao Lin , Bhuwan Dhingra

Deep learning models develop successive representations of their input in sequential layers, the last of which maps the final representation to the output. Here we investigate the informational content of these representations by observing…

Computer Vision and Pattern Recognition · Computer Science 2023-02-28 Benjamin L. Badger

With the attention mechanism, transformers achieve significant empirical successes. Despite the intuitive understanding that transformers perform relational inference over long sequences to produce desirable representations, we lack a…

Machine Learning · Computer Science 2024-04-02 Yufeng Zhang , Boyi Liu , Qi Cai , Lingxiao Wang , Zhaoran Wang

Chain-of-thought (CoT) prompting assumes that generated reasoning reflects a model's internal computation. We show this assumption is wrong in a specific, measurable way: models internally detect their own reasoning errors but outwardly…

Computation and Language · Computer Science 2026-05-12 Aojie Yuan , Zhiyuan Julian Su , Haiyue Zhang , Yi Nian , Yue Zhao

Concept bottleneck models (CBMs) are a class of interpretable neural network models that predict the target response of a given input based on its high-level concepts. Unlike the standard end-to-end models, CBMs enable domain experts to…

Machine Learning · Computer Science 2023-07-04 Sungbin Shin , Yohan Jo , Sungsoo Ahn , Namhoon Lee

Existing approaches for analyzing neural network activations, such as PCA and sparse autoencoders, rely on strong structural assumptions. Generative models offer an alternative: they can uncover structure without such assumptions and act as…

Machine Learning · Computer Science 2026-02-09 Grace Luo , Jiahai Feng , Trevor Darrell , Alec Radford , Jacob Steinhardt

Large Language Models often achieve strong performance by generating long intermediate chain-of-thought reasoning. However, it remains unclear when a model's final answer is actually determined during generation. If the answer is already…

Computation and Language · Computer Science 2026-04-27 Ayan Datta , Zhixue Zhao , Bhuvanesh Verma , Radhika Mamidi , Mounika Marreddy , Alexander Mehler

We propose a principle for exploring context in machine learning models. Starting with a simple assumption that each observation may or may not depend on its context, a conditional probability distribution is decomposed into two parts:…

Machine Learning · Computer Science 2019-01-23 Yun Zeng