Canada

Fine-tuning has been proven to be a simple and effective technique to transfer the learned knowledge of Pre-trained Language Models (PLMs) to downstream tasks. However, vanilla fine-tuning easily overfits the target data and degrades the generalization ability. Most existing studies attribute it to catastrophic forgetting, and they retain the pre-trained knowledge indiscriminately without identifying what knowledge is transferable. Motivated by this, we frame fine-tuning into a causal graph and discover that the crux of catastrophic forgetting lies in the missing causal effects from the pre-trained data. Based on the causal view, we propose a unified objective for fine-tuning to retrieve the causality back. Intriguingly, the unified objective can be seen as the sum of the vanilla fine-tuning objective, which learns new knowledge from target data, and the causal objective, which preserves old knowledge from PLMs. Therefore, our method is flexible and can mitigate negative transfer while preserving knowledge. Since endowing models with commonsense is a long-standing challenge, we implement our method on commonsense QA with a proposed heuristic estimation to verify its effectiveness. In the experiments, our method outperforms state-of-the-art fine-tuning methods on all six commonsense QA datasets and can be implemented as a plug-in module to inflate the performance of existing QA models.

&lt;iframe src=&quot;https://app.sli.do/event/or9aypsi961h595NQtD7kk/embed/polls/55827ba6-4c6d-4dcc-aa8a-9f7a0f00cdaa&quot; width=&quot;300&quot; height=&quot;400&quot;&gt;&lt;/iframe&gt;

ACL 2023

Preserving Commonsense Knowledge from Pre-trained Language Models via Causal Inference

pre-trained language models

causal inference

fine-tuning

## Welcome to EMNLP 2024! 
We are excited to welcome you to one of the most prominent conferences in the field of Natural Language Processing. This year, EMNLP 2024 is being held in a hybrid format,
offering both virtual and in-person participation in beautiful Miami. Due to a record-breaking number of submissions, we've expanded the total number of accepted papers to accommodate more cutting-edge research from around the globe.
### [Conference Handbook](https://drive.google.com/file/d/1WPROgxjLAC96AJL7Ugy0tEnYm7dkrbHt/view?usp=sharing)<br><br><br>

You are required to register for this event. <br>**Please register [here](https://2024.emnlp.org/registration/).**<br> The EMNLP 2024 event page on Underline will be open to public one week prior to the event.

Please register!

EMNLP 2024

EMNLP 2024 will take place in Miami, Florida from Nov 12th to Nov 16th, 2024, at the Hyatt Regency Miami Hote and on Underline for remote participants.

Fine-tuning has been proven to be a simple and effective technique to transfer the learned knowledge of Pre-trained Language Models (PLMs) to downstream tasks. However, vanilla fine-tuning easily overfits the target data and degrades the generalization ability. Most existing studies attribute it to catastrophic forgetting, and they retain the pre-trained knowledge indiscriminately without identifying what knowledge is transferable. Motivated by this, we frame fine-tuning into a causal graph and discover that the crux of catastrophic forgetting lies in the missing causal effects from the pre-trained data. Based on the causal view, we propose a unified objective for fine-tuning to retrieve the causality back. Intriguingly, the unified objective can be seen as the sum of the vanilla fine-tuning objective, which learns new knowledge from target data, and the causal objective, which preserves old knowledge from PLMs. Therefore, our method is flexible and can mitigate negative transfer while preserving knowledge. Since endowing models with commonsense is a long-standing challenge, we implement our method on commonsense QA with a proposed heuristic estimation to verify its effectiveness. In the experiments, our method outperforms state-of-the-art fine-tuning methods on all six commonsense QA datasets and can be implemented as a plug-in module to inflate the performance of existing QA models.

<iframe src="https://app.sli.do/event/or9aypsi961h595NQtD7kk/embed/polls/55827ba6-4c6d-4dcc-aa8a-9f7a0f00cdaa" width="300" height="400"></iframe>

technical paper

### Welcome to ACL 2023, the 61st Annual Meeting of the Association for Computational Linguistics! 
<br>The conference will be held in Toronto, Canada, July 9-14, 2023.<br><br>
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.<br>
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)**
<br><br><br><br>

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.

Zero-shot transfer learning for Dialogue State Tracking (DST) helps to handle a variety of task-oriented dialogue domains without the cost of collecting in-domain data. Existing works mainly study common data- or model-level augmentation methods to enhance the generalization but fail to effectively decouple semantics of samples, limiting the zero-shot performance of DST. In this paper, we present a simple and effective ``divide, conquer and combine'' solution, which explicitly disentangles the semantics of seen data, and leverages the performance and robustness with the mixture-of-experts mechanism. Specifically, we divide the seen data into semantically independent subsets and train corresponding experts, the newly unseen samples are mapped and inferred with mixture-of-experts with our designed ensemble inference.Extensive experiments on MultiWOZ2.1 upon T5-Adapter show our schema significantly and consistently improves the zero-shot performance, achieving the SOTA on settings without external knowledge, with only 10M trainable parameters.

<iframe src="https://app.sli.do/event/8PaBoicsuHK7beXFunvrtr/embed/polls/c026a1a2-09f3-41f2-b0f2-0687f20286de" width="300" height="400"></iframe>

Divide, Conquer, and Combine: Mixture of Semantic-Independent Experts for Zero-Shot Dialogue State Tracking

In this paper, we present AfricaPOS, the largest part-of-speech (POS) dataset for 20 typologically diverse African languages. We discuss the challenges in annotating POS for these languages using the universal dependencies (UD) guidelines. We conducted extensive POS baseline experiments using both conditional random field and several multilingual pre-trained language models. We applied various cross-lingual transfer models trained with data available in the UD. Evaluating on the AfricaPOS dataset, we show that choosing the best transfer language(s) in both single-source and multi-source setups greatly improves the POS tagging performance of the target languages, in particular when combined with parameter-fine-tuning methods. Crucially, transferring knowledge from a language that matches the language family and morphosyntactic properties seems to be more effective for POS tagging in unseen languages.

<iframe src="https://app.sli.do/event/8PaBoicsuHK7beXFunvrtr/embed/polls/50a5f01b-4f68-4313-b5f1-0c5fc7689617" width="300" height="400"></iframe>

MasakhaPOS: Part-of-Speech Tagging for Typologically Diverse African languages

Many open-domain dialogue models pre-trained with social media comments can generate coherent replies but have difficulties producing engaging responses. This phenomenon might mainly result from the deficiency of annotated human-human conversations and the misalignment with human preference. In this paper, we propose a novel and efficient framework Diamante to boost the open-domain chatbot, where two kinds of human feedback (including explicit demonstration and implicit preference) are collected and leveraged. By asking annotators to select or amend the model-generated candidate responses, Diamante efficiently collects the human demonstrated responses and constructs a Chinese chit-chat dataset. To enhance the alignment with human preference, Diamante leverages the implicit preference in the data collection process and introduces the generation-evaluation joint training. Comprehensive experiments indicate that the Diamante dataset and joint training paradigm can significantly boost the performance of pre-trained dialogue models. The overall engagingness of the previous state-of-the-art model has been improved remarkably by 50% in Chinese open-domain conversations.

<iframe src="https://app.sli.do/event/or9aypsi961h595NQtD7kk/embed/polls/21fee97f-2ffb-4993-b478-9778a7666598" width="300" height="400"></iframe>

Towards Boosting the Open-Domain Chatbot with Human Feedback

Question-Answering (QA) has seen significant advances recently, achieving near human-level performance over some benchmarks. However, these advances focus on high-resourced languages such as English, while the task remains unexplored for most other languages, mainly due to the lack of annotated datasets. This work presents a native QA dataset for an East African language, Tigrinya. The dataset contains 10.6K question-answer pairs spanning 572 paragraphs extracted from 290 news articles on various topics. The dataset construction method is discussed, which is applicable to constructing similar resources for related languages. We present comprehensive experiments and analyses of several resource-efficient approaches to QA, including monolingual, cross-lingual, and multilingual setups, along with comparisons against machine-translated silver data. Our strong baseline models reach 76% in the F1 score, while the estimated human performance is 92%, indicating that the benchmark presents a good challenge for future work. We make the dataset, models, and leaderboard publicly available.

<iframe src="https://app.sli.do/event/kgQkh7cPzt9VEM52HG3MJu/embed/polls/d2c4f58e-9457-4d57-a1e5-da2d14818c47" width="300" height="400"></iframe>

Question-Answering in a Low-resourced Language: Benchmark Dataset and Models for Tigrinya

Despite the surprising few-shot performance of in-context learning (ICL), it is still a common practice to randomly sample examples to serve as context. This paper advocates a new principle for ICL: self-adaptive in-context learning. The self-adaption mechanism is introduced to help each sample find an in-context example organization (i.e., selection and permutation) that can derive the correct prediction, thus maximizing performance. To validate the effectiveness of self-adaptive ICL, we propose a general select-then-rank framework and instantiate it with new selection and ranking algorithms. Upon extensive evaluation on eight different NLP datasets, our self-adaptive ICL method achieves a 40% relative improvement over the common practice setting. Further analysis reveals the enormous potential of self-adaptive ICL that it might be able to close the gap between ICL and finetuning given more advanced algorithms. Our code will be released to facilitate future research.

<iframe src="https://app.sli.do/event/kgQkh7cPzt9VEM52HG3MJu/embed/polls/10651e02-818e-4257-8c8e-f19d64c156e7" width="300" height="400"></iframe>

Self-Adaptive In-Context Learning: An Information Compression Perspective for In-Context Example Selection and Ordering

In-context learning, where pre-trained language models learn to perform tasks from task examples and instructions in their contexts, has attracted much attention in the NLP community. However, the ability of in-context learning is not fully exploited because language models are not explicitly trained to learn in context. To this end, we propose PICL (Pre-training for In-Context Learning), a framework to enhance the language models' in-context learning ability by pre-training the model on a large collection of "intrinsic tasks" in the general plain-text corpus using the simple language modeling objective. PICL encourages the model to infer and perform tasks by conditioning on the contexts while maintaining task generalization of pre-trained models. We evaluate the in-context learning performance of the model trained with PICL on seven widely-used text classification datasets and the Super-NaturalInstrctions benchmark, which contains 100+ NLP tasks formulated to text generation. Our experiments show that PICL is more effective and task-generalizable than a range of baselines, outperforming larger language models with nearly 4x parameters. The code is publicly available at https://github.com/thu-coai/PICL.

<iframe src="https://app.sli.do/event/eX4PwSmcCc1EvCqjxcYdKB/embed/polls/844eccb7-db55-4e51-b118-c9d63ffa7560" width="300" height="400"></iframe>

Pre-Training to Learn in Context

Large language models (LMs) beyond a certain scale, demonstrate the emergent capability of generating free-text rationales for their predictions via chain-of-thought (CoT) prompting.While CoT can yield dramatically improved performance, such gains are only observed for sufficiently large LMs. Even more concerning, there is little guarantee that the generated rationales are consistent with LM's predictions or faithfully justify the decisions. In this work, we propose SCOTT, a faithful knowledge distillation method to learn a small, self-consistent CoT model from a teacher model that is orders of magnitude larger. To form better supervision, we elicit rationales supporting the gold answers from a large LM (teacher) by contrastive decoding, which encourages the teacher to generate tokens that become more plausible only when the answer is considered. To ensure faithful distillation, we use the teacher-generated rationales to learn a student LM with a counterfactual reasoning objective, which prevents the student from ignoring the rationales to make inconsistent predictions. Experiments show that while yielding comparable performance, our method leads to a more faithful model than baselines. Further analysis shows that such a model respects the rationales more when making decisions; thus, we can improve its performance more by refining its rationales.

SCOTT: Self-Consistent Chain-of-Thought Distillation

Selective rationales and counterfactual examples have emerged as two effective, complementary classes of interpretability methods for analyzing and training NLP models. However, prior work has not explored how these methods can be integrated to combine their complementary advantages. We overcome this limitation by introducing CREST (ContRastive Edits with Sparse raTionalization), a joint framework for selective rationalization and counterfactual text generation, and show that this framework leads to improvements in counterfactual quality, model robustness, and interpretability. First, CREST generates valid counterfactuals that are more natural than those produced by previous methods, and subsequently can be used for data augmentation at scale, reducing the need for human-generated examples. Second, we introduce a new loss function that leverages CREST counterfactuals to regularize selective rationales and show that this regularization improves both model robustness and rationale quality, compared to methods that do not leverage CREST counterfactuals. Our results demonstrate that CREST successfully bridges the gap between selective rationales and counterfactual examples, addressing the limitations of existing methods and providing a more comprehensive view of a model's predictions.

<iframe src="https://app.sli.do/event/6qSnTPZvioRLfFJpZUjhCw/embed/polls/5cc7afb1-4a0c-4403-9939-b72cfa1d517c" width="300" height="400"></iframe>

CREST: A Joint Framework for Rationalization and Counterfactual Text Generation

Tracking characters and locations throughout a story can help improve the understanding of its plot structure. Prior research has analyzed characters and locations from text independently without grounding characters to their locations in narrative time. Here, we address this gap by proposing a new spatial relationship categorization task. The objective of the task is to assign a spatial relationship category for every character and location co-mention within a window of text, taking into consideration linguistic context, narrative tense, and temporal scope. To this end, we annotate spatial relationships in approximately $2500$ book excerpts and train a model using contextual embeddings as features to predict these relationships. When applied to a set of books, this model allows us to test several hypotheses on mobility and domestic space, revealing that protagonists are more mobile than non-central characters and that women as characters tend to occupy more interior space than men. Overall, our work is the first step towards joint modeling and analysis of characters and places in narrative text.

<iframe src="https://app.sli.do/event/6qSnTPZvioRLfFJpZUjhCw/embed/polls/4a8f746d-7f73-46ad-a875-177a88c618b1" width="300" height="400"></iframe>

Grounding Characters and Places in Narrative Text

Understanding interpersonal communication requires, in part, understanding the social context and norms in which a message is said. However, current methods for identifying offensive content in such communication largely operate independent of context, with only a few approaches considering community norms or prior conversation as context. Here, we introduce a new approach to identifying inappropriate communication by explicitly modeling the social relationship between the individuals. We introduce a new dataset of contextually-situated judgments of appropriateness and show that large language models can readily incorporate relationship information to accurately identify appropriateness in a given context. Using data from online conversations and movie dialogues, we provide insight into how the relationships themselves function as implicit norms and quantify the degree to which context-sensitivity is needed in different conversation settings. Further, we also demonstrate that contextual-appropriateness judgments are predictive of other social factors expressed in language such as condescension and politeness.

<iframe src="https://app.sli.do/event/532fnPQGAZvefKDC1fZsuL/embed/polls/d153b43f-ccef-4eef-a243-37c61441d7bc" width="300" height="400"></iframe>

Downloads

Next from ACL 2023

Divide, Conquer, and Combine: Mixture of Semantic-Independent Experts for Zero-Shot Dialogue State Tracking

Similar lecture

Does Fine-Tuning LLMs on New Knowledge Encourage Hallucinations?