201 citations · 214 across the 9 of their papers we have counts for
7 papers · 1 filter
Where Should a Document Live: Context, Representations, or Parameters?
Nathanaël Carraz Rakotonirina, Momchil Hardalov, Gonzalo Iglesias +1
To answer questions outside of their pre-training data, large language models (LLMs) need access to new information, which can be presented in the context window as documents, enco…
Correct, Concise and Complete: Multi-stage Training For Adaptive Reasoning
Nathanaël Carraz Rakotonirina, Ren Pang, Neha Anna John +2
The reasoning capabilities of large language models (LLMs) have improved substantially through increased test-time computation, typically in the form of intermediate tokens known a…
From Tools to Teammates: Evaluating LLMs in Multi-Session Coding Interactions
Nathanaël Carraz Rakotonirina, Mohammed Hamdy, Jon Ander Campos +5
Large Language Models (LLMs) are increasingly used in working environments for a wide range of tasks, excelling at solving individual problems in isolation. However, are they also…
Evil twins are not that evil: Qualitative insights into machine-generated prompts
Nathanaël Carraz Rakotonirina, Corentin Kervadec, Francesca Franzon +1
It has been widely observed that language models (LMs) respond in predictable ways to algorithmically generated prompts that are seemingly unintelligible. This is both a sign that…
MemoryPrompt: A Light Wrapper to Improve Context Tracking in Pre-trained Language Models
Nathanaël Carraz Rakotonirina, Marco Baroni
Transformer-based language models (LMs) track contextual information through large, hard-coded input windows. We introduce MemoryPrompt, a leaner approach in which the LM is comple…
Towards hate speech detection in low-resource languages: Comparing ASR to acoustic word embeddings on Wolof and Swahili
Christiaan Jacobs, Nathanaël Carraz Rakotonirina, Everlyn Asiko Chimoto +2
We consider hate speech detection through keyword spotting on radio broadcasts. One approach is to build an automatic speech recognition (ASR) system for the target low-resource la…
Can discrete information extraction prompts generalize across language models?
Nathanaël Carraz Rakotonirina, Roberto Dessì, Fabio Petroni +2
We study whether automatically-induced prompts that effectively extract information from a language model can also be used, out-of-the-box, to probe other language models for the s…