Analogical reasoning is a powerful inductive mechanism, widely used in human cognition and increasingly applied in artificial intelligence. Formal frameworks for analogical inference have been developed for Boolean domains, where inference is provably sound for affine functions and approximately correct for functions close to affine. These results have informed the design of analogy-based classifiers. However, they do not extend to regression tasks or continuous domains. In this paper, we revisit analogical inference from a foundational perspective. We first present a counterexample showing that existing generalization bounds fail even in the Boolean setting. We then introduce a unified framework for analogical reasoning in real-valued domains based on parameterized analogies defined via generalized means. This model subsumes both Boolean classification and regression, and supports analogical inference over continuous functions. We characterize the class of analogy-preserving functions in this setting and derive both worst-case and average-case error bounds under smoothness assumptions. Our results offer a general theory of analogical inference across discrete and continuous domains.

The undeniable revolution brought forth by Large Language Models (LLMs) stems from the amazing fluency of the texts they generate, mastering language with seemingly human-like finesse. This fluency raises a key scientific question: How much lexical knowledge do LLMs actually capture in order to produce such fluent language? To address this, we present ALF, a freely available, analogical dataset endowed with rich lexicographic information grounded in Meaning-Text Theory for the French language. It comprises 2600 fine-grained lexical analogies with which we evaluate the lexical ability of five off-the-shelf LLMs, namely ChatGPT-4o mini, Llama3.0-8B, Llama3.1-8B, Qwen2.5-14B, and Mistral7B. Their performance spans from 45% for Mistral, through about 55% for the ChatGPT and Llama models, and up to nearly 60% for Qwen2.5-14B, thus qualifying ALF as a challenging dataset. Experimenting with larger models (OpenAI o1, Llama3.0/3.1-70B, and Qwen2.5-32B) yields rather limited returns considering the drastic increase in computational cost. We further identify certain types of analogies and prompting methods that reveal performance disparities.

Liyan Wang, Haotong Wang, Yves Lepage. Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2025.

Retrieval-Augmented Neural Machine Translation (RANMT), which augments translation models with relevant examples fetched by a similarity retriever, is proficient in well-resourced translations. However, its inherent advantages are not fully realized in low-resource contexts, where the sparsity of data can often result in less relevant or less useful information for translation. Recent literature indicates that nearest-neighbor examples from small training data have the unfortunate effect of impairing RANMT performance. Our examination of 16 low-resource tasks reveals a sharp deterioration in performance of a multilingual language model when conditioned on retrieved examples compared to direct translation. To address this problem, we explore a framework based on analogical reasoning, aiming to enhance the capacity of language models to infer translations from parallel examples in limited data settings. This framework mimics a cognitive process of human translation by structuring examples in analogy patterns. We propose a multi-objective learning strategy that augments vanilla training for conditional translation to learn latent knowledge from examples. We also investigate different retrieval methods for selecting translation examples based on lexical similarity, semantic relatedness, or a combination of both. The results show that our approach is effective in optimizing RANMT in low-resource settings, delivering notable improvements across all retrieval settings. In particular, augmented training akin to reasoning with analogies in two directions, contributes significantly to deriving benefits from examples, even when their relevance is limited. Moreover, our approach demonstrates superior performance in low-resource translation tasks compared to prompting large language models in few-shot contexts. It also proves to be competitive with models that have been extensively trained using substantial amounts of supervised data.

Mixup, an interpolation-based method that implicitly generates synthetic examples for training, has shown effectiveness in tasks such as image and text classification. Standard mixup randomly interpolates two samples of images and their labels. In this article, we apply mixup to low-resource machine translation tasks by interpolating in the hidden space. We investigate the impact of different mixing coefficients on this technique. We also explore whether semantically related or unrelated samples provide more benefits for interpolation compared to random selection. To investigate this, we extend the standard mixup approach by selecting samples based on distance and experimenting with different sampling settings. Our experiments are conducted across several low-resource language pairs, including Lower Sorbian and Upper Sorbian, Lower Sorbian and German, and Upper Sorbian and German. Through systematic experiments on multiple language pairs, we evaluate the effectiveness of mixup data augmentation in improving low-resource machine translation performance. Our findings indicate that the standard mixup technique enhances the quality of machine translation, resulting in an average increase of 1.9 BLEU points over the baseline Transformer model. The choice of mixing coefficients has minimal impact on translation quality, which suggests that fine-tuning these coefficients is not essential to benefit from mixup. In addition, the standard mixup performs robustly, as selecting either the most similar or most dissimilar samples for mixing does not provide a significant improvement over it.

This paper introduces Continued Pre-training on Analogies (CPoA) to incorporate pre-trained language models with analogical abilities, aiming at improving performance in low-resource translations without data augmentation. We continue training the models on sentence analogies retrieved from a translation corpus. Considering the sparsity of analogy in corpora, especially in low-resource scenarios, we propose exploring approximate analogies between sentences. We attempt to find sentence analogies that might not conform to formal criteria for entire sentences but partial pieces. When training the models, we introduce a weighting scalar pertaining to the quality of analogies to adjust the influence: emphasizing closer analogies while diminishing the impact of far ones. We evaluate our approach on a low-resource translation task: German-Upper Sorbian. The results show that CPoA using 10 times fewer instances can effectively attain gains of +1.4 and +1.3 BLEU points over the original model in two translation directions. This improvement is more pronounced when there are fewer parallel examples.

Morphological generation is a task where given a lemma and a morphosyntactic description of the target form, we are asked to generate the target form. Knowing that the syntactic and semantic relations to other forms are reflected by the word form itself, we show how to exploit these relations between word forms, holistically, that is, as a whole, to derive the target form without even breaking them into morphemes. Experimental results show that by organising the lexica into analogical grids we are able to improve the accuracy of morphological generation by up to 8% in low data scenarios. Our holistic approach always performs better than a morpheme-based baseline. We also enquire possible improvements by using data augmentation for neural approaches, especially in low data scenarios. However, our system seems not to gain any advantage from having more data after some point in time.

This paper presents an approach to enhance the quality of machine translation by leveraging middle sentences as pivot points and employing dual reinforcement learning. Conventional methods for generating parallel sentence pairs for machine translation rely on parallel corpora, which may be scarce, resulting in limitations in translation quality. In contrast, our proposed method entails training two machine translation models in opposite directions, utilizing the middle sentence as a bridge for a virtuous feedback loop between the two models. This feedback loop resembles reinforcement learning, facilitating the models to make informed decisions based on mutual feedback. Experimental results substantiate that our proposed method significantly improves machine translation quality.

This article studies the application of the #BenderRule in Natural Language Processing (NLP) articles according to two dimensions. Firstly, in a contrastive manner, by considering two major international conferences, LREC and ACL, and secondly, in a diachronic manner, by inspecting nearly 14,000 articles over a period of time ranging from 2000 to 2020 for LREC and from 1979 to 2020 for ACL. For this purpose, we created a corpus from LREC and ACL articles from the above-mentioned periods, from which we manually annotated nearly 1,000. We then developed two classifiers to automatically annotate the rest of the corpus. Our results show that LREC articles tend to respect the #BenderRule (80 to 90% of them respect it), whereas 30 to 40% of ACL articles do not. Interestingly, over the considered periods, the results appear to be stable for the two conferences, even though a rebound in ACL 2020 could be a sign of the influence of the blog post about the #BenderRule.

Morphological segmentation is useful for processing Mongolian. In this paper, we manually build a morphological segmentation data set for Mongolian. We then present a character-based encoder-decoder model with attention mechanism to perform the morphological segmentation task. We further investigate the influence of analogy features extracted from scratch and improve the performance of our model using multi languages setting. Experimental results show that our encoder-decoder model with attention mechanism provides a strong baseline for Mongolian morphological segmentation. The analogy features provide useful information to the model and improve the performance of the system. The use of multi languages data set shows the capability of our model to acquire knowledge through different languages and delivers the best result.