AIConfPaper
← Search

Maxime Peyrard

14 accepted papers

2026

Interactive Evaluation of Large Language Models for Multi-Requirement Software Engineering Tasks

Dimitrios Rontogiannis, Maxime Peyrard, Nicolas Baldwin, Martin Josifoski, Robert West, Dimitrios Gunopulos
AAAI 2026technical

Standard single-turn, static benchmarks fall short in evaluating the nuanced capabilities of Large Language Models (LLMs) on complex tasks such as software engineering. In this work, we propose a novel interactive evaluation framework that assesses LLMs on multi-requirement programming tasks through

Cited by 0Source ↗PDF ↗Scholar ↗
2026

Mechanistic Interpretability as Statistical Estimation: A Variance Analysis

Maxime Méloux, François Portet, Maxime Peyrard
ICML 2026poster

Mechanistic Interpretability (MI) aims to reverse-engineer model behaviors by identifying functional sub-networks. Yet, the scientific validity of these findings depends on their stability. In this work, we argue that circuit discovery is not a standalone task but a statistical estimation problem bu…

Cited by 0Source ↗Scholar ↗
2025

Date Fragments: A Hidden Bottleneck of Tokenization for Temporal Reasoning

Gagan Bhatia, Maxime Peyrard, Wei Zhao
EMNLP 2025

Modern BPE tokenisers often split calendar dates into meaningless fragments, e.g., “20250312” → “202”, “503”, “12”, inflating token counts and obscuring the inherent structure needed for robust temporal reasoning. In this work, we (1) introduce a simple yet interpretable metric, termed date fragment

Cited by 0Source ↗PDF ↗Scholar ↗
2025

Everything, Everywhere, All at Once: Is Mechanistic Interpretability Identifiable?

Maxime Méloux, Silviu Maniu, François Portet, Maxime Peyrard
ICLR 2025poster

As AI systems are increasingly deployed in high-stakes applications, ensuring their interpretability is essential. Mechanistic Interpretability (MI) aims to reverse-engineer neural networks by extracting human-understandable algorithms embedded within their structures to explain their behavior. This…

Cited by 0Source ↗PDF ↗code ↗Scholar ↗
2025

zip2zip: Inference-Time Adaptive Tokenization via Online Compression

Saibo Geng, Nathan Ranchin, Yunzhen Yao, Maxime Peyrard, Chris Wendler, Michael Gastpar +1
NeurIPS 2025poster

Tokenization efficiency plays a critical role in the performance and cost of large language models (LLMs), yet most models rely on static tokenizers optimized on general-purpose corpora. These tokenizers’ fixed vocabularies often fail to adapt to domain- or language-specific inputs, leading to longe…

Cited by 0Source ↗code ↗Scholar ↗
2024

A Glitch in the Matrix? Locating and Detecting Language Model Grounding with Fakepedia

Giovanni Monea, Maxime Peyrard, Martin Josifoski, Vishrav Chaudhary, Jason Eisner, Emre Kiciman +3
ACL 2024long

Large language models (LLMs) have an impressive ability to draw on novel information supplied in their context. Yet the mechanisms underlying this contextual grounding remain unknown, especially in situations where contextual information contradicts factual knowledge stored in the parameters, which…

Cited by 10Source ↗PDF ↗code ↗Scholar ↗
2023

Exploiting Asymmetry for Synthetic Training Data Generation: SynthIE and the Case of Information Extraction

Martin Josifoski, Marija Sakota, Maxime Peyrard, Robert West
EMNLP 2023long main

Large language models (LLMs) have great potential for synthetic data generation. This work shows that useful data can be synthetically generated even for tasks that cannot be solved directly by LLMs: for problems with structured outputs, it is possible to prompt an LLM to perform the task in the rev…

Cited by 0Source ↗code ↗Scholar ↗
2023

Grammar-Constrained Decoding for Structured NLP Tasks without Finetuning

Saibo Geng, Martin Josifoski, Maxime Peyrard, Robert West
EMNLP 2023long main

Despite their impressive performance, large language models (LMs) still struggle with reliably generating complex output structures when not finetuned to follow the required output format exactly. To address this issue, grammar-constrained decoding (GCD) can be used to control the generation of LMs,…

Cited by 0Source ↗code ↗Scholar ↗
2022

GenIE: Generative Information Extraction

Martin Josifoski, Nicola De Cao, Maxime Peyrard, Fabio Petroni, Robert West
NAACL 2022long

Structured and grounded representation of text is typically formalized by closed information extraction, the problem of extracting an exhaustive set of (subject, relation, object) triplets that are consistent with a predefined set of entities and relations from a knowledge base schema. Most existing…

Cited by 84Source ↗PDF ↗code ↗Scholar ↗
2022

Invariant Language Modeling

Maxime Peyrard, Sarvjeet Ghotra, Martin Josifoski, Vidhan Agarwal, Barun Patra, Dean Carignan +3
EMNLP 2022main

Modern pretrained language models are critical components of NLP pipelines. Yet, they suffer from spurious correlations, poor out-of-domain generalization, and biases.Inspired by recent progress in causal machine learning, in particular the invariant risk minimization (IRM) paradigm, we propose inva…

Cited by 19Source ↗PDF ↗code ↗Scholar ↗
2022

Predicting Is Not Understanding: Recognizing and Addressing Underspecification in Machine Learning

Damien Teney, Maxime Peyrard, Ehsan Abbasnejad
ECCV 2022poster

"Machine learning models are typically designed for maximum accuracy on validation data. This predictive criterion rarely captures all desirable properties, in particular how a model matches a domain expert’s \emph{understanding} of the task. In this situation, known as underspecification, two model…

Cited by 33Source ↗PDF ↗Scholar ↗
2021

Laughing Heads: Can Transformers Detect What Makes a Sentence Funny?

Maxime Peyrard, Beatriz Borges, Kristina Gligorić, Robert West
IJCAI 2021poster

The automatic detection of humor poses a grand challenge for natural language processing. Transformer-based systems have recently achieved remarkable results on this task, but they usually (1) were evaluated in setups where serious vs humorous texts came from entirely different sources, and (2) f…

Cited by 15Source ↗PDF ↗code ↗Scholar ↗