English
Related papers

Related papers: The Attentional White Bear Effect in Transformer L…

200 papers

Despite the increasing prevalence of large language models (LLMs), we still have a limited understanding of how their representational spaces are structured. This limits our ability to interpret how and what they learn or relate them to…

Exposure bias describes the phenomenon that a language model trained under the teacher forcing schema may perform poorly at the inference stage when its predictions are conditioned on its previous predictions unseen from the training…

Computation and Language · Computer Science 2020-04-02 Yifan Xu , Kening Zhang , Haoyu Dong , Yuezhou Sun , Wenlong Zhao , Zhuowen Tu

Neural language models exhibit impressive performance on a variety of tasks, but their internal reasoning may be difficult to understand. Prior art aims to uncover meaningful properties within model representations via probes, but it is…

Computation and Language · Computer Science 2021-09-21 Mycal Tucker , Peng Qian , Roger Levy

Exposure bias has been regarded as a central problem for auto-regressive language models (LM). It claims that teacher forcing would cause the test-time generation to be incrementally distorted due to the training-generation discrepancy.…

Machine Learning · Computer Science 2021-09-06 Tianxing He , Jingzhao Zhang , Zhiming Zhou , James Glass

Recovering meaningful concepts from language model activations is a central aim of interpretability. While existing feature extraction methods aim to identify concepts that are independent directions, it is unclear if this assumption can…

Interpretability benefits the theoretical understanding of representations. Existing word embeddings are generally dense representations. Hence, the meaning of latent dimensions is difficult to interpret. This makes word embeddings like a…

Computation and Language · Computer Science 2023-06-27 Minxue Xia , Hao Zhu

We investigate the internal structure of language model computations using causal analysis and demonstrate two motifs: (1) a form of adaptive computation where ablations of one attention layer of a language model cause another layer to…

Machine Learning · Computer Science 2023-08-01 Thomas McGrath , Matthew Rahtz , Janos Kramar , Vladimir Mikulik , Shane Legg

Common methods for interpreting neural models in natural language processing typically examine either their structure or their behavior, but not both. We propose a methodology grounded in the theory of causal mediation analysis for…

Computation and Language · Computer Science 2020-11-24 Jesse Vig , Sebastian Gehrmann , Yonatan Belinkov , Sharon Qian , Daniel Nevo , Simas Sakenis , Jason Huang , Yaron Singer , Stuart Shieber

Self-supervised learning has shown great success in Speech Recognition. However, it has been observed that finetuning all layers of the learned model leads to lower performance compared to resetting top layers. This phenomenon is attributed…

Computation and Language · Computer Science 2024-05-15 Valentin Vielzeuf

Standard imitation learning can fail when the expert demonstrators have different sensory inputs than the imitating agent. This is because partial observability gives rise to hidden confounders in the causal graph. In previous work, to work…

Machine Learning · Computer Science 2024-08-27 Risto Vuorio , Pim de Haan , Johann Brehmer , Hanno Ackermann , Daniel Dijkman , Taco Cohen

Authorship analysis is an important subject in the field of natural language processing. It allows the detection of the most likely writer of articles, news, books, or messages. This technique has multiple uses in tasks related to…

Recently multimodal transformer models have gained popularity because their performance on language and vision tasks suggest they learn rich visual-linguistic representations. Focusing on zero-shot image retrieval tasks, we study three…

Computation and Language · Computer Science 2021-02-02 Lisa Anne Hendricks , John Mellor , Rosalia Schneider , Jean-Baptiste Alayrac , Aida Nematzadeh

Learning causal representations from observational and interventional data in the absence of known ground-truth graph structures necessitates implicit latent causal representation learning. Implicit learning of causal mechanisms typically…

Machine Learning · Computer Science 2024-08-19 Shayan Shirahmad Gale Bagi , Zahra Gharaee , Oliver Schulte , Mark Crowley

The powerful modeling capabilities of all-attention-based transformer architectures often cause overfitting and - for natural language processing tasks - lead to an implicitly learned internal language model in the autoregressive…

Machine Learning · Computer Science 2022-09-21 Timo Lohrenz , Björn Möller , Zhengyang Li , Tim Fingscheidt

Human intelligence relies in part on our brains' ability to create abstract mental models that succinctly capture the hidden blueprint of our reality. Such abstract world models notably allow us to rapidly navigate novel situations by…

Artificial Intelligence · Computer Science 2023-12-12 Quentin RV. Ferry , Joshua Ching , Takashi Kawai

Surprisal theory links human processing effort to the predictability of an upcoming linguistic unit, but empirical work often leaves the notion of a unit underspecified. In practice, experimental stimuli are segmented into linguistically…

Computation and Language · Computer Science 2026-05-01 Samuel Kiegeland , Vésteinn Snæbjarnarson , Tim Vieira , Ryan Cotterell

A foundational principle in cognitive science holds that intelligent agents do not learn by storing experiences as isolated instances, but by forming abstract schemas that capture relational structure shared across situations. Even though…

Machine Learning · Computer Science 2026-05-26 Elnaz Rahmati , Nona Ghazizadeh , Zhivar Sourati , Nina Rouhani , Morteza Dehghani

Representations derived from models such as BERT (Bidirectional Encoder Representations from Transformers) and HuBERT (Hidden units BERT), have helped to achieve state-of-the-art performance in dimensional speech emotion recognition.…

Sound · Computer Science 2023-12-29 Vikramjit Mitra , Jingping Nie , Erdrin Azemi

Despite known differences between reading and listening in the brain, recent work has shown that text-based language models predict both text-evoked and speech-evoked brain activity to an impressive degree. This poses the question of what…

Computation and Language · Computer Science 2024-06-18 Subba Reddy Oota , Emin Çelik , Fatma Deniz , Mariya Toneva

Even though large language models (LLMs) have demonstrated remarkable capability in solving various natural language tasks, the capability of an LLM to follow human instructions is still a concern. Recent works have shown great improvements…

Computation and Language · Computer Science 2024-03-05 Xinbo Wu , Lav R. Varshney