中文
相关论文

相关论文: Counterfactual reasoning: an analysis of in-contex…

200 篇论文

Present language understanding methods have demonstrated extraordinary ability of recognizing patterns in texts via machine learning. However, existing methods indiscriminately use the recognized patterns in the testing phase that is…

计算与语言 · 计算机科学 2021-06-08 Fuli Feng , Jizhi Zhang , Xiangnan He , Hanwang Zhang , Tat-Seng Chua

Using more test-time computation during language model inference, such as generating more intermediate thoughts or sampling multiple candidate answers, has proven effective in significantly improving model performance. This paper takes an…

机器学习 · 计算机科学 2025-08-20 Xingwu Chen , Miao Lu , Beining Wu , Difan Zou

Recent work has shown that Transformers can perform in-context learning for linear regression under restrictive assumptions, including i.i.d. data, Gaussian noise, and Gaussian regression coefficients. However, real-world data often violate…

机器学习 · 计算机科学 2026-03-20 Hoang T. H. Cao , Hai D. V. Trinh , Tho Quan , Lan V. Truong

In this paper, we describe an approach for modelling causal reasoning in natural language by detecting counterfactuals in text using multi-head self-attention weights. We use pre-trained transformer models to extract contextual embeddings…

计算与语言 · 计算机科学 2020-06-02 Rajaswa Patil , Veeky Baths

Scaling large language models (LLMs) leads to an emergent capacity to learn in-context from example demonstrations. Despite progress, theoretical understanding of this phenomenon remains limited. We argue that in-context learning relies on…

计算与语言 · 计算机科学 2023-03-15 Michael Hahn , Navin Goyal

Counterfactual reasoning has emerged as a crucial technique for generalizing the reasoning capabilities of large language models (LLMs). By generating and analyzing counterfactual scenarios, researchers can assess the adaptability and…

人工智能 · 计算机科学 2026-02-17 Shuai Yang , Qi Yang , Luoxi Tang , Yuqiao Meng , Nancy Guo , Jeremy Blackburn , Zhaohan Xi

Despite the increasing effectiveness of language models, their reasoning capabilities remain underdeveloped. In particular, causal reasoning through counterfactual question answering is lacking. This work aims to bridge this gap. We first…

计算与语言 · 计算机科学 2025-03-18 Alihan Hüyük , Xinnuo Xu , Jacqueline Maasch , Aditya V. Nori , Javier González

Large language models have shown tremendous performance in a variety of tasks. In-context learning -- the ability to improve at a task after being provided with a number of demonstrations -- is seen as one of the main contributors to their…

计算与语言 · 计算机科学 2023-05-23 Julian Coda-Forno , Marcel Binz , Zeynep Akata , Matthew Botvinick , Jane X. Wang , Eric Schulz

Probing and enhancing large language models' reasoning capacity remains a crucial open question. Here we re-purpose the reverse dictionary task as a case study to probe LLMs' capacity for conceptual inference. We use in-context learning to…

计算与语言 · 计算机科学 2024-02-27 Ningyu Xu , Qi Zhang , Menghan Zhang , Peng Qian , Xuanjing Huang

Transformer tends to overallocate attention to irrelevant context. In this work, we introduce Diff Transformer, which amplifies attention to the relevant context while canceling noise. Specifically, the differential attention mechanism…

计算与语言 · 计算机科学 2025-04-08 Tianzhu Ye , Li Dong , Yuqing Xia , Yutao Sun , Yi Zhu , Gao Huang , Furu Wei

Despite the advanced capabilities of large language models (LLMs), their temporal reasoning ability remains underdeveloped. Prior works have highlighted this limitation, particularly in maintaining temporal consistency when understanding…

计算与语言 · 计算机科学 2025-06-18 Jongho Kim , Seung-won Hwang

Evaluating hypothetical statements about how the world would be had a different course of action been taken is arguably one key capability expected from modern AI systems. Counterfactual reasoning underpins discussions in fairness, the…

机器学习 · 计算机科学 2022-10-04 Kevin Xia , Yushu Pan , Elias Bareinboim

In-context learning refers to the ability of a model to condition on a prompt sequence consisting of in-context examples (input-output pairs corresponding to some task) along with a new query input, and generate the corresponding output.…

计算与语言 · 计算机科学 2023-08-15 Shivam Garg , Dimitris Tsipras , Percy Liang , Gregory Valiant

Conventional audio classification relied on predefined classes, lacking the ability to learn from free-form text. Recent methods unlock learning joint audio-text embeddings from raw audio-text pairs describing audio in natural language.…

多媒体 · 计算机科学 2024-01-11 Ali Vosoughi , Luca Bondi , Ho-Hsiang Wu , Chenliang Xu

The impressive performance of recent language models across a wide range of tasks suggests that they possess a degree of abstract reasoning skills. Are these skills general and transferable, or specialized to specific tasks seen during…

计算与语言 · 计算机科学 2024-04-01 Zhaofeng Wu , Linlu Qiu , Alexis Ross , Ekin Akyürek , Boyuan Chen , Bailin Wang , Najoung Kim , Jacob Andreas , Yoon Kim

Transformers have demonstrated remarkable performance in natural language processing and related domains, as they largely focus on sequential, autoregressive next-token prediction tasks. Yet, they struggle in logical reasoning, not…

人工智能 · 计算机科学 2025-10-08 Renee Ge , Qianli Liao , Tomaso Poggio

Counterfactual reasoning is widely recognized as one of the most challenging and intricate aspects of causality in artificial intelligence. In this paper, we evaluate the performance of large language models (LLMs) in counterfactual…

计算与语言 · 计算机科学 2026-04-14 Yuefei Chen , Vivek K. Singh , Jing Ma , Ruixiang Tang

We propose an architecture for training generative models of counterfactual conditionals of the form, 'can we modify event A to cause B instead of C?', motivated by applications in robot control. Using an 'adversarial training' paradigm, an…

机器人学 · 计算机科学 2020-09-23 Simón C. Smith , Subramanian Ramamoorthy

We recently reported evidence that large language models are capable of solving a wide range of text-based analogy problems in a zero-shot manner, indicating the presence of an emergent capacity for analogical reasoning. Two recent…

计算与语言 · 计算机科学 2024-05-01 Taylor Webb , Keith J. Holyoak , Hongjing Lu

Recently, strong results have been demonstrated by Deep Recurrent Neural Networks on natural language transduction problems. In this paper we explore the representational power of these models using synthetic grammars designed to exhibit…

神经与进化计算 · 计算机科学 2015-11-04 Edward Grefenstette , Karl Moritz Hermann , Mustafa Suleyman , Phil Blunsom
‹ 上一页 1 2 3 10 下一页 ›