Korea, Republic of

Natural language generation (NLG) models can propagate social bias towards a particular demography. Though several studies investigated bias from data and models, the NLG task distinctively uses stochastic decoders that can positively or negatively impact the bias-sensitive tokens initially predicted by the model. To address this gap in research, we present an extensive analysis of bias from decoding techniques for open-domain language generation considering the entire decoding space. We analyze to what extent bias metrics like toxicity and sentiment are impacted by the individual components of decoder algorithms. We also analyze the trade-off between bias scores and human-annotated generation quality throughout the decoder space. Together, these methods reveal the imperative of testing inference time bias and provide evidence on the usefulness of inspecting the entire decoding spectrum.

COLING 2022

Quantifying Bias from Decoding Techniques in Natural Language Generation

toxicity

language model

bias

### Welcome to The 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025)

Message from the General Chair: 
*It is my great pleasure and honor to welcome you to the 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025), held in beautiful Vienna, Austria, from July 27 to August 1, 2025. ACL2025continues our field’s tradition of excellence in scholarship, innovation, and inclusivity, and I am deeply grateful to the many volunteers who have worked tirelessly to bring this event to life.* 
[Read more](https://drive.google.com/file/d/1GI_hvOpjswAuYdUTromfeDiPpCcqidwg/view?usp=sharing)

To access this event page, you need to log in with the **email address you registered with**. Access credentials will be sent to your email from Underline - subject line "Welcome to ACL 2025". Please be sure to check your spam email folder if you do not see an email confirmation right away.

Please log in

To access this event page, you need to log in with the **email address you registered with**. 

ACL 2025

Welcome to The 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025)

poster

[![](https://assets.underline.io/uploads/markdown_image/1/image/81a3c0317d24f663b49b996875024d45.png)](https://aclanthology.org/volumes/2022.coling-1/)



### THE CONFERENCE WE KNOW AND WE WANT.
**COLING**, the International Conference on Computational Linguistics, is one of the premier conferences for the natural language processing and computational linguistics.

First established in 1965, the biennial COLING conference is held in diverse parts of the globe and attracts participants from both top-ranked research centers and emerging countries. Today, the most important developments in our field are taking place not only in universities and academic research institutes but also in industrial research departments including tech-startups. COLING provides opportunities for all these communities to showcase their exciting discovery.

In fall of 2022, COLING will be held in Gyeongju in a hybrid format. All participants can either present at the venue site or join virtually. As more people get vaccinated, we are happy to provide safer environments for our colleagues. We believe that COLING 2022 will be one of the conferences, free from the pandemic. The hybrid format gives presenters and sponsors a valuable opportunity to promote their companies in both an online and in-person venue. For the first time in a long time, customers can interact with their sponsor's products first-hand. The online venue, too, gives sponsors the chance to network with those unable to attend the in-person session.

The hybrid format ultimately lends itself to greater exposure for our sponsors: COLING2022 will let you reach more potential partners and customers than ever before!

To gain access to this event page you are required to register. Find more information and pay the registration fee on event the organizer’s website **[https://coling2022.org/reg](https://coling2022.org/reg)**

COLING, the International Conference on Computational Linguistics, is one of the premier conferences for the natural language processing and computational linguistics.

The zero-shot cross-lingual ability of models pretrained on multilingual and even monolingual corpora has spurred many hypotheses to explain this intriguing empirical result. However, due to the costs of pretraining, most research uses public models whose pretraining methodology, such as the choice of tokenization, corpus size, and computational budget, might differ drastically. When researchers pretrain their own models, they often do so under a constrained budget, and the resulting models might underperform significantly compared to SOTA models. These experimental differences led to various inconsistent conclusions about the nature of the cross-lingual ability of these models. To help further research on the topic, we released 10 monolingual byte-level models rigorously pretrained under the same configuration with a large compute budget (equivalent to 420 days on a V100) and corpora that are 4 times larger than the original BERT's. Because they are tokenizer-free, the problem of unseen token embeddings is eliminated, thus allowing researchers to try a wider range of cross-lingual experiments in languages with different scripts. Additionally, we release two models pretrained on non-natural language texts that can be used in sanity-check experiments. Experiments on QA and NLI tasks show that our monolingual models achieve competitive performance to the multilingual one, and hence can be served to strengthen our understanding of cross-lingual transferability in language models. Models are available at: https://huggingface.co/monobyte.

MonoByte: A Pool of Monolingual Byte-level Language Models

This paper considers the task of parsing low-resource languages in a scenario where parallel English data and also a limited seed of annotated sentences in the target language are available, as for example in bootstrapping parallel treebanks.
We focus on constituency parsing using Role and Reference Grammar (RRG), a theory that has so far been understudied in computational linguistics but that is widely used in typological research, i.e., in particular in the context of low-resource languages. Starting from an existing RRG parser, we propose two strategies for low-resource parsing: first, we extend the parsing model into a cross-lingual parser, exploiting the parallel data in the high-resource language and unsupervised word alignments by providing internal states of the source-language parser to the target-language parser. Second, we adopt self-training, thereby iteratively expanding the training data, starting from the seed, by including the most
confident new parses in each round.
Both in simulated scenarios and with a real low-resource language (Daakaka), we find substantial and complementary improvements from both self-training and cross-lingual parsing. Moreover, we also experimented with using gloss embeddings in addition to token embeddings in the target language, and this also improves results. Finally, starting from what we have for Daakaka, we also consider parsing a related language (Dalkalaen) where glosses and English translations are available but no annotated trees at all, i.e., a no-resource scenario wrt.~syntactic annotations. We start with cross-lingual parser trained on Daakaka with glosses and use self-training to adapt it to Dalkalaen. The results are surprisingly good.

Improving Low-resource RRG Parsing with Cross-lingual Self-training

Target-oriented dialog aims to reach a global target through multi-turn conversation. The key to the task is the global planning towards the target, which flexibly guides the dialog concerning the context. However, existing target-oriented dialog works take a local and greedy strategy for response generation, where global planning is absent. In this work, we propose global planning for target-oriented dialog on a commonsense knowledge graph (KG). We design a global reinforcement learning with the planned paths to flexibly adjust the local response generation model towards the global target. We also propose a KG-based method to collect target-oriented samples automatically from the chit-chat corpus for model training. Experiments show that our method can reach the target with a higher success rate, fewer turns, and more coherent responses.

TopKG: Target-oriented Dialog via Global Planning on Knowledge Graph

Previous work has demonstrated that pre-trained large language models (LLM) acquire knowledge during pre-training which enables reasoning over relationships between words (e.g, hyponymy) and more complex inferences over larger units of meaning such as sentences. Here, we investigate whether lexical entailment (LE, i.e. hyponymy or the is a relation between words) can be generalised in a compositional manner. Accordingly, we introduce PLANE (Phrase-Level Adjective-Noun Entailment), a new benchmark to test models on fine-grained compositional entailment using adjective-noun phrases. Our experiments show that knowledge extracted via In–Context and transfer learning is not enough to solve PLANE. However, a LLM trained on PLANE can generalise well to out–of–distribution sets, since the required knowledge can be stored in the representations of subwords (SW) tokens.

Testing Large Language Models on Compositionality and Inference with Phrase-Level Adjective-Noun Entailment

Although it is widely agreed that world knowledge plays a significant role in quantifier scope disambiguation (QSD), there has been only very limited work on how to integrate this knowledge into a QSD model. This study contributes to this scarce line of research by incorporating into a machine learning model our knowledge about relations, as conveyed by a manageable closed class of function
words: prepositions. For data, we use a scope-disambiguated corpus created by AnderBois, Brasoveanu and Henderson, which is additionally annotated with prepositional senses using Schneider et al’s Semantic Network of Adposition and Case Supersenses (SNACS) scheme. By applying Manshadi and Allen’s method to
the corpus, we were able to inspect the information gain provided by prepositions for the QSD task. Statistical analysis of the performance of the classifiers, trained in scenarios with and without preposition information, supports the claim that prepositional senses have a strong positive impact on the learnability of automatic
QSD systems.


Prepositions Matter in Quantifier Scope Disambiguation

Most attempts on Text-to-SQL task using encoder-decoder approach show a big problem of dramatic decline in performance for new databases. For the popular Spider dataset, despite models achieving 70% accuracy on its development or test sets, the same models show a huge decline below 20% accuracy for unseen databases. The root causes for this problem are complex and they cannot be easily fixed by adding more manually created training. We propose a solution that is a hybrid system using automated training-data augmentation. Our system consists of a rule-based and a deep learning components that extract crucial information to predict correct SQL, leading to double-double accuracy improvement.

Addressing Limitations of Encoder-Decoder Based Approach to Text-to-SQL

Existing studies on the analysis of persuasion in online discussions focus on investigating the effectiveness of comments in discussions and ignore the analysis of the effectiveness of debaters over multiple discussions.
In this paper, we propose to quantify debaters effectiveness in the online discussion platform: ``ChangeMyView'' in order to explore diverse insights into their persuasion strategies. In particular, targeting debaters with different levels of effectiveness (e.g., good vs. poor), various behavioral characteristics (e.g., engagement) and text stylistic features (e.g., used frames) of debaters are carefully examined, leading to several outcomes that can be the backbone of writing assistants and persuasive text generation.

Analyzing Persuasion Strategies of Debaters on Social Media

Current practices in building new NLP models for low-resourced languages rely either on Machine Translation of training sets from better resourced languages or on cross-lingual transfer from them. Still we can see a considerable performance gap between the models originally trained within better resourced languages and the models transferred from them. In this study we test the possibility of (1) using natural annotation to build synthetic training sets from resources not initially designed for the target downstream task and (2) employing curriculum learning methods to select the most suitable examples from synthetic training sets. We test this hypothesis across seven Slavic languages and across three curriculum learning strategies on Named Entity Recognition as the downstream task. We also test the possibility of fine-tuning the synthetic resources to reflect linguistic properties, such as the grammatical case and gender, both of which are important for the Slavic languages. We demonstrate the possibility to achieve the mean F1 score of 0.78 across the three basic entities types for Belarusian starting from zero resources in comparison to the baseline of 0.63 using the zero-shot transfer from English. For comparison, the English model trained on the original set achieves the mean F1-score of 0.75.

Applying Natural Annotation and Curriculum Learning to Named Entity Recognition for Under-Resourced Languages

Diversity can be decomposed into three dis-
tinct concepts, namely: variety, balance and dis-
parity. This paper borrows from the extensive
formalization and measures of diversity devel-
oped in ecology in order to evaluate the variety
and balance of multiword expression annota-
tion produced by automatic annotation systems.
The measures of richness, normalized richness,
and two variations of Hill’s evenness are con-
sidered in this paper. We observe how these
measures behave against increasingly smaller
samples of gold annotations of multiword ex-
pressions and use their comportment to validate
or invalidate their pertinence for multiword ex-
pressions in annotated texts. We apply the vali-
dated measures to annotations in 14 languages
produced by systems during the PARSEME
shared task on automatic identification of mul-
tiword expressions and on the gold versions of
the corpora. We also explore the limits of such
evaluation by studying the impact of lemmati-
zation errors in the Turkish corpus used in the
shared task.

Evaluating Diversity of Multiword Expressions in Annotated Text

Discourse parsing has proven to be useful for a number of NLP tasks that require complex reasoning. However, over a decade since the advent of the Penn Discourse Treebank, predicting implicit discourse relations in text remains challenging. There are several possible reasons for this, and we hypothesize that models should be exposed to more context as it plays an important role in accurate human annotation; meanwhile adding uncertainty measures can improve model accuracy and calibration. To thoroughly investigate this phenomenon, we perform a series of experiments to determine 1) the effects of context on human judgments, and 2) the effect of quantifying uncertainty with annotator confidence ratings on model accuracy and calibration (which we measure using the Brier score (Brier, 1950)). We find that including annotator accuracy and confidence improves model accuracy, and incorporating confidence in the model's temperature function can lead to models with significantly better-calibrated confidence measures. We also find some insightful qualitative results regarding human and model behavior on these datasets.

Premium content

Downloads

Next from COLING 2022

MonoByte: A Pool of Monolingual Byte-level Language Models

Similar lecture

Unmasking Style Sensitivity: A Causal Analysis of Bias Evaluation Instability in Large Language Models