English
Related papers

Related papers: ExeDec: Execution Decomposition for Compositional …

200 papers

Several studies have reported the inability of Transformer models to generalize compositionally, a key type of generalization in many NLP tasks such as semantic parsing. In this paper we explore the design space of Transformer models…

Artificial Intelligence · Computer Science 2022-03-04 Santiago Ontañón , Joshua Ainslie , Vaclav Cvicek , Zachary Fisher

We consider the problem of neural semantic parsing, which translates natural language questions into executable SQL queries. We introduce a new mechanism, execution guidance, to leverage the semantics of SQL. It detects and excludes faulty…

Computation and Language · Computer Science 2018-09-17 Chenglong Wang , Kedar Tatwawadi , Marc Brockschmidt , Po-Sen Huang , Yi Mao , Oleksandr Polozov , Rishabh Singh

Compositional reinforcement learning is a promising approach for training policies to perform complex long-horizon tasks. Typically, a high-level task is decomposed into a sequence of subtasks and a separate policy is trained to perform…

Machine Learning · Computer Science 2023-06-09 Kishor Jothimurugan , Steve Hsu , Osbert Bastani , Rajeev Alur

Synthesizing a program that realizes a logical specification is a classical problem in computer science. We examine a particular type of program synthesis, where the objective is to synthesize a strategy that reacts to a potentially…

Artificial Intelligence · Computer Science 2020-01-01 Alberto Camacho , Sheila A. McIlraith

Systematic generalization remains challenging for current language models, which are known to be both sensitive to semantically similar permutations of the input and to struggle with known concepts presented in novel contexts. Although…

Computation and Language · Computer Science 2025-05-28 Sondre Wold , Lucas Georges Gabriel Charpentier , Étienne Simon

We introduce DafnyCOMP, a benchmark for evaluating large language models (LLMs) on compositional specification generation in Dafny. Unlike prior benchmarks that focus on single-function tasks, DafnyCOMP targets programs composed of multiple…

Programming Languages · Computer Science 2025-09-30 Xu Xu , Xin Li , Xingwei Qu , Jie Fu , Binhang Yuan

Chain-of-Thought (CoT) training has markedly advanced the reasoning capabilities of large language models (LLMs), yet the mechanisms by which CoT training enhances generalization remain inadequately understood. In this work, we demonstrate…

Machine Learning · Computer Science 2026-02-13 Xinhao Yao , Ruifeng Ren , Yun Liao , Lizhong Ding , Yong Liu

Presently with technology node scaling, an accurate prediction model at early design stages can significantly reduce the design cycle. Especially during logic synthesis, predicting cell congestion due to improper logic combination can…

Machine Learning · Computer Science 2021-11-12 Amur Ghose , Vincent Zhang , Yingxue Zhang , Dong Li , Wulong Liu , Mark Coates

Large language models (LLMs) exhibit remarkable task generalization, solving tasks they were never explicitly trained on with only a few demonstrations. This raises a fundamental question: When can learning from a small set of tasks…

Machine Learning · Computer Science 2025-06-10 Amirhesam Abedsoltan , Huaqing Zhang , Kaiyue Wen , Hongzhou Lin , Jingzhao Zhang , Mikhail Belkin

The rise of large-scale multimodal models has paved the pathway for groundbreaking advances in generative modeling and reasoning, unlocking transformative applications in a variety of complex tasks. However, a pressing question that remains…

Computation and Language · Computer Science 2024-04-19 Semih Yagcioglu , Osman Batur İnce , Aykut Erdem , Erkut Erdem , Desmond Elliott , Deniz Yuret

In-context learning has shown great success in i.i.d semantic parsing splits, where the training and test sets are drawn from the same distribution. In this setup, models are typically prompted with demonstrations that are similar to the…

Computation and Language · Computer Science 2023-06-27 Itay Levy , Ben Bogin , Jonathan Berant

The advent of sensor networks presents untapped opportunities for synthesis. We examine the problem of synthesis of behavioral specifications into networks of programmable sensor blocks. The particular behavioral specification we consider…

Other Computer Science · Computer Science 2011-11-09 Ryan Mannion , Harry Hsieh , Susan Cotterell , Frank Vahid

Instruction tuning has emerged as the key in aligning large language models (LLMs) with specific task instructions, thereby mitigating the discrepancy between the next-token prediction objective and users' actual goals. To reduce the labor…

Computation and Language · Computer Science 2024-04-10 Zifeng Wang , Chun-Liang Li , Vincent Perot , Long T. Le , Jin Miao , Zizhao Zhang , Chen-Yu Lee , Tomas Pfister

In text-to-SQL tasks -- as in much of NLP -- compositional generalization is a major challenge: neural networks struggle with compositional generalization where training and test distributions differ. However, most recent attempts to…

Computation and Language · Computer Science 2022-05-05 Yujian Gan , Xinyun Chen , Qiuping Huang , Matthew Purver

Human linguistic capacity is often characterized by compositionality and the generalization it enables -- human learners can produce and comprehend novel complex expressions by composing known parts. Several benchmarks exploit…

Computation and Language · Computer Science 2022-12-22 Najoung Kim , Tal Linzen , Paul Smolensky

The effectiveness of large language models (LLMs) is often hindered by duplicated data in their extensive pre-training datasets. Current approaches primarily focus on detecting and removing duplicates, which risks the loss of valuable…

Computation and Language · Computer Science 2024-07-10 Nan He , Weichen Xiong , Hanwen Liu , Yi Liao , Lei Ding , Kai Zhang , Guohua Tang , Xiao Han , Wei Yang

Program synthesis is the task of automatically generating a program consistent with a given specification. A natural way to specify programs is to provide examples of desired input-output behavior, and many current program synthesis…

Machine Learning · Computer Science 2020-07-28 Alexander Suh , Yuval Timen

Code execution is a fundamental aspect of programming language semantics that reflects the exact behavior of the code. However, most pre-trained models for code intelligence ignore the execution trace and only rely on source code and…

Programming Languages · Computer Science 2023-05-10 Chenxiao Liu , Shuai Lu , Weizhu Chen , Daxin Jiang , Alexey Svyatkovskiy , Shengyu Fu , Neel Sundaresan , Nan Duan

We focus on prediction problems with structured outputs that are subject to output validity constraints, e.g. pseudocode-to-code translation where the code must compile. While labeled input-output pairs are expensive to obtain, "unlabeled"…

Machine Learning · Computer Science 2023-10-26 Sang Michael Xie , Tengyu Ma , Percy Liang

Humans are highly efficient learners, with the ability to grasp the meaning of a new concept from just a few examples. Unlike popular computer vision systems, humans can flexibly leverage the compositional structure of the visual world,…

Computer Vision and Pattern Recognition · Computer Science 2021-05-21 Yanli Zhou , Brenden M. Lake