Papers by Emmanuel Chemla
It is a Bird Therefore it is a Robin: On BERT’s Internal Consistency Between Hypernym Knowledge and Logical Words (2023.findings-acl)
Copied to clipboard
| Challenge: | Existing models for lexical knowledge of NLP systems only use collocational cues for questions . lexicon of these systems is limited to content words and logical words. |
| Approach: | They propose a method to test the understanding of the hypernymy relationship by measuring its antisymmetry according to the models. |
| Outcome: | The proposed model shows that it is weak on the semantic tasks of the Masked Language Modeling task. |
Biasless Language Models Learn Unnaturally: How LLMs Fail to Distinguish the Possible from the Impossible (2026.eacl-long)
Copied to clipboard
| Challenge: | linguists have discovered patterns which hold across virtually all known natural languages . lingulists are able to learn languages by comparing their learning curves to those of humans . |
| Approach: | They compare LLM learning curves on existing and "impossible" datasets . they find that GPT-2 learns each language and its impossible counterpart equally easily . |
| Outcome: | The proposed model learns each language and its impossible counterpart equally easily, the study shows . the study also shows that the proposed model does not provide any kind of separation between the possible and the impossible . |
Bridging the Empirical-Theoretical Gap in Neural Network Formal Language Learning Using Minimum Description Length (2024.acl-long)
Copied to clipboard
| Challenge: | Neural networks offer good approximation to many tasks but fail to reach perfect generalization. |
| Approach: | They propose to use a formal language to test whether a theoretically correct solution is not an optimum of commonly used objectives. |
| Outcome: | The proposed model fails to reach the theoretically correct solution even with regularization techniques. |
Improving Spoken Language Modeling with Phoneme Classification: A Simple Fine-tuning Approach (2024.emnlp-main)
Copied to clipboard
| Challenge: | Generating speech through a pipeline that operates at the text level typically loses nuances, intonations, and non-verbal vocalizations. |
| Approach: | They show that fine-tuning speech representation models on phoneme classification leads to more context-invariant representations, and language models trained on these units achieve comparable lexical comprehension to ones trained on hundred times more data. |
| Outcome: | Recent advances in speech representation modeling have shown that learning language directly from speech is feasible. |
Minimum Description Length Recurrent Neural Networks (2022.tacl-1)
Copied to clipboard
| Challenge: | We show that neural networks that optimize a minimum description length score master memory challenges and perform addition with 100% accuracy. |
| Approach: | They train neural networks to optimize a Minimum Description Length score . they show that they master tasks involving memory challenges and perform addition . |
| Outcome: | The proposed models master languages and perform addition with 100% accuracy . they show that they can generalize from small training corpora and large training corpus . |
On the Spontaneous Emergence of Discrete and Compositional Signals (2020.acl-main)
Copied to clipboard
| Challenge: | Using a continuous latent space, we are able to train using backpropagation and show discrete messages nevertheless naturally emerge. |
| Approach: | They propose a general framework to study language emergence through signaling games with neural agents. |
| Outcome: | The proposed framework shows that discrete messages naturally emerge and that they are not compositional. |