Canada

How to usefully encode compositional task structure has long been a core challenge in AI. Recent work in chain of thought prompting has shown that for very large neural language models (LMs), explicitly demonstrating the inferential steps involved in a target task may improve performance over end-to-end learning that focuses on the target task alone. However, chain of thought prompting has significant limitations due to its dependency on huge pretrained LMs. In this work, we present compositional fine-tuning (CFT): an approach based on explicitly decomposing a target task into component tasks, and then fine-tuning smaller LMs on a curriculum of such component tasks. We apply CFT to recommendation tasks in two domains, world travel and local dining, as well as a previously studied inferential task (sports understanding). We show that CFT outperforms end-to-end learning even with equal amounts of data, and gets consistently better as more component tasks are modeled via fine-tuning. Compared with chain of thought prompting, CFT performs at least as well using LMs only 7.4% of the size, and is moreover applicable to task domains for which data are not available during pretraining.

ACL 2023

Learning to Perform Complex Tasks through Compositional Fine-Tuning of Language Models

language model fine-tuning

machine reasoning

multi-step reasoning

large language models

curriculum learning

## Welcome!
"I am excited to welcome you to this year’s edition of the Conference on Empirical Methods in Natural Language Processing! Importantly, it marks the 30th edition of EMNLP. With over 8,000 submissions, more than 3,000 accepted papers, and thousands of attendees, we have come a long way from that first
workshop, which had 14 accepted papers. As the field looks ahead, Suzhou is the fitting location for celebrating this milestone: rooted in a long literary tradition, yet modern and forward-looking, and home to a large share of the NLP community." 

*Message from the General Chair, Dirk Hovy*

EMNLP 2025

Celebrate 30 Years of EMNLP! 
EMNLP 2025 will be held in Suzhou, China from November 5th to November 9th, 2025.

workshop paper

### Welcome to ACL 2023, the 61st Annual Meeting of the Association for Computational Linguistics! 
 The conference will be held in Toronto, Canada, July 9-14, 2023. 
Following the succession of the recent conferences in our field, ACL 2023 will adopt a hybrid format.
While the impact of Covid has considerably diminished in terms of traveling, obtaining visas to Canada
entails a very long process. Moreover, the global economic conditions pose challenges for many individuals to travel to conferences. Recognizing these circumstances, we know many participants may not be
able to attend the conference in person. Therefore, we are committed to providing a great virtual platform
so everyone has the opportunity to interact with other participants and enjoy the conference. Based on the
current registered participants, approxiately 30% have chosen to attend the conference virtually. Whether
you join us in person or virtually, we sincerely hope everyone has a remarkable conference experience. 
This General Chair’s message is where I express my gratitude to the many individuals who have made
enormous contributions to the conference over the past year.

Read [**ACL 2023 General Chair's message**](https://docs.google.com/document/d/1WobYM7norbG4dI48s75HfJoD89qgX5a_F-6U8AteLSA/edit?usp=sharing/) in full.

##### **[Conference Handbook](https://2023.aclweb.org/downloads/acl2023-handbook.pdf)**

The Association for Computational Linguistics (ACL) is the premier international scientific and professional society for people working on computational problems involving human language, a field often referred to as either computational linguistics or natural language processing.

Recent work has shown that prompting language models with code-like representations of natural language leads to performance improvements on structured reasoning tasks. However, such tasks comprise only a small subset of all natural language tasks. In our work, we seek to answer whether or not code-prompting is the preferred way of interacting with language models in general. We compare code and text prompts across three popular GPT models (davinci, code-davinci-002, and text-davinci-002) on a broader selection of tasks (e.g., QA, sentiment, summarization) and find that with few exceptions, code prompts do not consistently outperform text prompts. Furthermore, we show that the style of code prompt has a large effect on performance for some but not all tasks and that fine-tuning on text instructions leads to better relative performance of code prompts.

Exploring the Curious Case of Code Prompts

Saliency maps can explain a neural model's predictions by identifying important input features. They are difficult to interpret for laypeople, especially for instances with many features. In order to make them more accessible, we formalize the underexplored task of translating saliency maps into natural language and compare methods that address two key challenges of this approach -- what and how to verbalize. In both automatic and human evaluation setups, using token-level attributions from text classification tasks, we compare two novel methods (search-based and instruction-based verbalizations) against conventional feature importance representations (heatmap visualizations and extractive rationales), measuring simulatability, faithfulness, helpfulness and ease of understanding. Instructing GPT-3.5 to generate saliency map verbalizations yields plausible explanations which include associations, abstractive summarization and commonsense reasoning, achieving by far the highest human ratings, but they are not faithfully capturing numeric information and are inconsistent in their interpretation of the task. In comparison, our search-based, model-free verbalization approach efficiently completes templated verbalizations, is faithful by design, but falls short in helpfulness and simulatability. Our results suggest that saliency map verbalization makes feature attribution explanations more comprehensible and less cognitively challenging to humans than conventional representations.

Saliency Map Verbalization: Comparing Feature Importance Representations from Model-free and Instruction-based Methods

The relationship between communicated language and intended meaning is often probabilistic and sensitive to context. Numerous strategies attempt to estimate such a mapping, often leveraging recursive Bayesian models of communication. In parallel, large language models (LLMs) have been increasingly applied to semantic parsing applications, tasked with inferring logical representations from natural language. While existing LLM explorations have been largely restricted to literal language use, in this work, we evaluate the capacity of LLMs to infer the meanings of pragmatic utterances. Specifically, we explore the case of threshold estimation on the gradable adjective "strong", contextually conditioned on a strength prior, then extended to composition with qualification, negation, polarity inversion, and class comparison. We find that LLMs can derive context-grounded, human-like distributions over the interpretations of several complex pragmatic utterances, yet struggle composing with negation. These results inform the inferential capacity of statistical language models, and their use in pragmatic and semantic parsing applications.

Evaluating statistical language models as pragmatic reasoners

Script learning studies how daily events unfold. Previous works tend to consider a script as a linear sequence of events while ignoring the potential branches that arise due to people's circumstantial choices. We hence propose Choice-75, the first benchmark that challenges intelligent systems to predict decisions given descriptive scenarios, containing 75 scripts and more than 600 scenarios. While large language models demonstrate overall decent performances, there is still notable room for improvement in many hard scenarios.

Choice-75: A Dataset on Decision Branching in Script Learning

Larger language models, such as GPT-3, have shown to be excellent in many tasks. However, we demonstrate that out-of-ordinary questions can throw the model off guard. This work focuses on finding answers to negated complementary ques- tions in commonsense scenarios. We illustrate how such questions adversely af- fect the model responses. We propose a model-agnostic methodology to improve the performance in negated complementary scenarios. Our method outperforms few-shot generation from GPT-3 (by more than 11 points) and, more importantly, highlights the significance of studying the response of large language models in negated complementary questions.

Negated Complementary Commonsense using Large Language Models

In this paper, we investigate whether symbolic semantic representations, extracted from deep semantic parsers, can help to reason over the states of involved entities in a procedural text. We consider a deep semantic parser~(TRIPS) and semantic role labeling as two sources of semantic parsing knowledge. 
First, we propose PROPOLIS, a symbolic parsing-based procedural reasoning framework.
Second, we integrate semantic parsing information into state-of-the-art neural models to conduct procedural reasoning.
Our experiments indicate that explicitly incorporating such semantic knowledge improves procedural understanding. This paper presents new metrics for evaluating procedural reasoning tasks that clarify the challenges and identify differences among neural, symbolic, and integrated models.

The Role of Semantic Parsing in Understanding Procedural Text

Large language models (LLMs) struggle on processing complicated observations in interactive decision making. To alleviate this issue, we propose a simple hierarchical prompting approach. Diverging from previous prompting approaches that always put the full observation (e.g. a web page) to the prompt, we propose to first construct an action-aware observation which is more condensed and relevant with a dedicated Summarizer prompt. The Actor prompt then predicts the next action based on the summarized history. While our method has broad applicability, we particularly demonstrate its efficacy in the complex domain of web navigation where a full observation often contains redundant and irrelevant information. Our approach outperforms the previous state-of-the-art prompting mechanism with the same LLM by 6.2% on task success rate, demonstrating its potential on interactive decision making tasks with long observation traces.

Hierarchical Prompting Assists Large Language Model on Web Navigation

Users' physical safety is an increasing concern as the market for intelligent systems continues to grow, where unconstrained systems may recommend users dangerous actions that can lead to serious injury. Covertly unsafe text is an area of particular interest, as such text may arise from everyday scenarios and are challenging to detect as harmful. We propose FARM, a novel framework leveraging external knowledge for trustworthy rationale generation in the context of safety. In particular, FARM foveates on missing knowledge to qualify the information required to reason in specific scenarios and retrieves this information with attribution to trustworthy sources. This knowledge is used to both classify the safety of the original text and generate human-interpretable rationales, shedding light on the risk of systems to specific user groups and helping both stakeholders manage the risks of their systems and policymakers to provide concrete safeguards for consumer safety. Our experiments show that FARM obtains state-of-the-art results on the SafeText dataset, showing absolute improvement in safety classification accuracy by 5.9%.

Foveate, Attribute, and Rationalize: Towards Physically Safe and Trustworthy AI

Large language models (LLMs) have achieved impressive performance on code
generation. However, for complex programming tasks, generating the correct
solution in one go becomes challenging, thus some prior works have designed
program repair approaches to improve code generation performance. In this work,
we propose Self-Debugging, which teaches a large language model to debug its
predicted program via few-shot demonstrations. In particular, we demonstrate that Self-Debugging can teach the large language model to perform rubber duck
debugging; i.e., without any feedback on the code correctness or error messages,
the model is able to identify its mistakes by explaining the generated code in
natural language. Self-Debugging achieves the state-of-the-art performance on
several code generation benchmarks, including the Spider dataset for text-to-SQL
generation, TransCoder for C++-to-Python translation, and MBPP for text-toPython generation. On the Spider benchmark where there are no unit tests to
verify the correctness of predictions, Self-Debugging with code explanation
consistently improves the baseline by 2−3%, and improves the prediction accuracy
on problems of the hardest label by 9%. On TransCoder and MBPP where unit tests
are available, Self-Debugging improves the baseline accuracy by up to 12%.
Meanwhile, by leveraging feedback messages and reusing failed predictions, Self-Debugging notably improves sample efficiency, and can match or outperform baseline models that generate more than 10× candidate programs.

Teaching Large Language Models to Self-Debug

Large Language Models (LLMs) are capable of performing zero-shot closed-book question answering tasks, based on their internal knowledge stored in parameters during pre-training. However, such internalized knowledge might be insufficient and incorrect, which could lead LLMs to generate factually wrong answers. Furthermore, fine-tuning LLMs to update their knowledge is expensive. To this end, we propose to augment the knowledge directly in the input of LLMs. Specifically, we first retrieve the relevant facts to the input question from the knowledge graph based on semantic similarities between the question and its associated facts. After that, we prepend the retrieved facts to the input question in the form of the prompt, which is then forwarded to LLMs to generate the answer. Our framework, Knowledge-Augmented language model PromptING (KAPING), requires no model training, thus completely zero-shot. We validate the performance of our KAPING framework on the knowledge graph question answering task, that aims to answer the user's question based on facts over a knowledge graph, on which ours outperforms relevant zero-shot baselines by up to 48% in average, across multiple LLMs of various sizes.

Next from ACL 2023

Exploring the Curious Case of Code Prompts

Similar lecture

Unleashing the Reasoning Potential of LLMs by Critique Fine-Tuning on One Problem

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES