paper-with-me

Papers

Memory in humans and deep language models: Linking hypotheses for model augmentation

2022-10-04 · Omri Raccah, Phoebe Chen, Ted L. Willke, David Poeppel, Vy A. Vo

The computational complexity of the self-attention mechanism in Transformer models significantly limits their ability to generalize over long temporal durations. Memory-augmentation, or the explicit storing of past information in external memory for subsequent predictions, has become a constructive avenue for mitigating this limitation. We argue that memory-augmented Transformers can benefit substantially from considering insights from the memory literature in humans. We detail an approach for integrating evidence from the human memory system through the specification of cross-domain linking hypotheses. We then provide an empirical demonstration to evaluate the use of surprisal as a linking hypothesis, and further identify the limitations of this approach to inform future research.

📄 PDF Abstract BibTeX arXiv:2210.01869

Code (0)

등록된 구현이 없습니다.

Methods 이 논문이 사용한 방법론

Attention 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Residual Connection 설명 없음
Label Smoothing Label Smoothing is a regularization technique that introduces noise for the labels. This accounts for the fact that datasets may have mistakes in them, so maximizing the…
Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
BPE Byte Pair Encoding, or BPE, is a subword segmentation algorithm that encodes rare and unknown words as sequences of subword units. The intuition is that various word…
Multi-Head Attention 설명 없음
Adam 설명 없음

Similar Papers 제목 키워드 기반

When transformers learn "impossible" languages, what do they learn?

2026-06-29 · Ram Janarthan, Coleman Haley, Sharon Goldwater arxiv

Recent work suggests that transformer language models show a bias towards human languages over unnatural ("impossible") languages argued to be unacquirable by humans. However, this literature has largely based these clai…

ComFact: A Benchmark for Linking Contextual Commonsense Knowledge

2022-10-23 · Silin Gao, Jena D. Hwang, Saya Kanno, Hiromi Wakaki 외

Understanding rich narratives, such as dialogues and stories, often requires natural language processing systems to access relevant knowledge from commonsense knowledge graphs. However, these systems typically retrieve f…

Knowledge GraphsResponse GenerationRetrieval

Disentangling the Roles of Curation, Data-Augmentation and the Prior in the Cold Posterior Effect

2021-06-11 · NeurIPS 2021 12 · Lorenzo Noci, Kevin Roth, Gregor Bachmann, Sebastian Nowozin 외

The "cold posterior effect" (CPE) in Bayesian deep learning describes the uncomforting observation that the predictive performance of Bayesian neural networks can be significantly improved if the Bayes posterior is artif…

Data Augmentation

Systematic Study of Long Tail Phenomena in Entity Linking

2018-08-01 · COLING 2018 8 · Filip Ilievski, Piek Vossen, Stefan Schlobach

State-of-the-art entity linkers achieve high accuracy scores with probabilistic methods. However, these scores should be considered in relation to the properties of the datasets they are evaluated on. Until now, there ha…

Entity Linking

Language models as tools for investigating the distinction between possible and impossible natural languages

2025-12-10 · Julie Kallini, Christopher Potts arxiv

We argue that language models (LMs) have strong potential as investigative tools for probing the distinction between possible and impossible natural languages and thus uncovering the inductive biases that support human l…