English
Related papers

Related papers: The Generalization Ridge: Information Flow in Natu…

200 papers

Retrieval-Augmented Generation (RAG) systems rely critically on the retriever module to surface relevant context for large language models. Although numerous retrievers have recently been proposed, each built on different ranking principles…

Information Retrieval · Computer Science 2026-02-26 Wenqing Zheng , Dmitri Kalaev , Noah Fatsi , Daniel Barcklow , Owen Reinert , Igor Melnyk , Senthil Kumar , C. Bayan Bruss

Large language models (LLMs) are often portrayed as merely imitating linguistic patterns without genuine understanding. We argue that recent findings in mechanistic interpretability (MI), the emerging field probing the inner workings of…

Computation and Language · Computer Science 2026-02-26 Pierre Beckmann , Matthieu Queloz

Retrieval-Augmented Generation (RAG) has emerged as a standard framework for knowledge-intensive NLP tasks, combining large language models (LLMs) with document retrieval from external corpora. Despite its widespread use, most RAG pipelines…

Information Retrieval · Computer Science 2025-08-26 Mandeep Rathee , V Venktesh , Sean MacAvaney , Avishek Anand

Retrieval-augmented generation (RAG) enables large language models (LLMs) to access external knowledge, helping mitigate hallucinations and enhance domain-specific expertise. Graph-based RAG enhances structural reasoning by introducing…

Computation and Language · Computer Science 2025-11-26 Linxiao Cao , Ruitao Wang , Jindong Li , Zhipeng Zhou , Menglin Yang

Learning Using Privileged Information is a particular type of knowledge distillation where the teacher model benefits from an additional data representation during training, called privileged information, improving the student model, which…

Computation and Language · Computer Science 2024-08-20 Rafael-Edy Menadil , Mariana-Iuliana Georgescu , Radu Tudor Ionescu

This paper proposes an information-theoretic representation learning framework, named conditional information flow maximization, to extract noise-invariant sufficient representations for the input data and target task. It promotes the…

Machine Learning · Computer Science 2024-08-13 Dou Hu , Lingwei Wei , Wei Zhou , Songlin Hu

Graph neural networks (GNNs), as the de-facto model class for representation learning on graphs, are built upon the multi-layer perceptrons (MLP) architecture with additional message passing layers to allow features to flow across nodes.…

Machine Learning · Computer Science 2023-08-07 Chenxiao Yang , Qitian Wu , Jiahua Wang , Junchi Yan

The effectiveness of Reinforcement Learning (RL) in Large Language Models (LLMs) depends on the nature and diversity of the data used before and during RL. In particular, reasoning problems can often be approached in multiple ways that rely…

Artificial Intelligence · Computer Science 2026-05-12 Aswin RRV , Jacob Dineen , Divij Handa , Mihir Parmar , Ben Zhou , Swaroop Mishra , Chitta Baral

Retrieval-Augmented Generation (RAG) enables large language models (LLMs) to access broader knowledge sources, yet factual inconsistencies persist due to noise in retrieved documents-even with advanced retrieval methods. We demonstrate that…

Computation and Language · Computer Science 2025-06-04 Yongjian Li , HaoCheng Chu , Yukun Yan , Zhenghao Liu , Shi Yu , Zheni Zeng , Ruobing Wang , Sen Song , Zhiyuan Liu , Maosong Sun

We present a hybrid statistical and grammar-based system for surface natural language generation (NLG) that uses grammar rules, conditions on using those grammar rules, and corpus statistics to determine the word order. We also describe how…

Computation and Language · Computer Science 2007-05-23 Adwait Ratnaparkhi

Retrieval-Augmented Generation (RAG) effectively mitigates hallucinations in LLMs by incorporating external knowledge. However, the inherent discrete representation of text in existing frameworks often results in a loss of semantic…

Computation and Language · Computer Science 2026-03-10 Pengcheng Zhou , Haochen Li , Zhiqiang Nie , JiaLe Chen , Qing Gong , Weizhen Zhang , Chun Yu

While compositional accounts of human language understanding are based on a hierarchical tree-like process, neural models like transformers lack a direct inductive bias for such tree structures. Introducing syntactic inductive biases could…

Computation and Language · Computer Science 2025-03-25 Ananjan Nandi , Christopher D. Manning , Shikhar Murty

Single-step retrieval-augmented generation (RAG) provides an efficient way to incorporate external information for simple question answering tasks but struggles with complex questions. Agentic RAG extends this paradigm by replacing…

Computation and Language · Computer Science 2026-05-08 Yijia Zheng , Marcel Worring

Natural language understanding (NLU) and Natural language generation (NLG) tasks hold a strong dual relationship, where NLU aims at predicting semantic labels based on natural language utterances and NLG does the opposite. The prior work…

Computation and Language · Computer Science 2020-10-16 Shang-Yu Su , Yung-Sung Chuang , Yun-Nung Chen

The Information Bottleneck (IB) principle offers a compelling theoretical framework to understand how neural networks (NNs) learn. However, its practical utility has been constrained by unresolved theoretical ambiguities and significant…

Machine Learning · Computer Science 2026-02-02 Charles Westphal , Stephen Hailes , Mirco Musolesi

Large language models (LLMs) are demonstrably capable of cross-lingual transfer, but can produce inconsistent output when prompted with the same queries written in different languages. To understand how language models are able to…

Computation and Language · Computer Science 2025-09-29 Zheng Wei Lim , Alham Fikri Aji , Trevor Cohn

Understanding how large language models (LLMs) represent natural language is a central challenge in natural language processing (NLP) research. Many existing methods extract word embeddings from an LLM, visualise the embedding space via…

Computation and Language · Computer Science 2026-01-12 Thomas Fabian

Prior work has found that pretrained language models (LMs) fine-tuned with different random seeds can achieve similar in-domain performance but generalize differently on tests of syntactic generalization. In this work, we show that, even…

Computation and Language · Computer Science 2024-06-06 Adithya Bhaskar , Dan Friedman , Danqi Chen

Grounded language models use external sources of information, such as knowledge graphs, to meet some of the general challenges associated with pre-training. By extending previous work on compositional generalization in semantic parsing, we…

Computation and Language · Computer Science 2024-06-10 Sondre Wold , Étienne Simon , Lucas Georges Gabriel Charpentier , Egor V. Kostylev , Erik Velldal , Lilja Øvrelid

Recently, unsupervised pre-training is gaining increasing popularity in the realm of computational linguistics, thanks to its surprising success in advancing natural language understanding (NLU) and the potential to effectively exploit…

Computation and Language · Computer Science 2019-11-15 Yuanxin Liu , Zheng Lin