1 citations · 2 across the 13 of their papers we have counts for
13 papers
Shieldstral
Antonia Calvi, Avinash Sooriyarachchi, Giada Pistilli +273
We introduce Shieldstral, a 3B-parameter policy-adaptive multimodal safety classifier that matches or outperforms models nearly 7 its size on text safety benchmarks and set…
Reasoning Arena: Trace Tournaments When Verifiable Rewards Fall Short
Han Zhou, Adam X. Yang, Laurence Aitchison +2
Reinforcement learning with verifiable rewards (RLVR) has become a leading paradigm for improving the reasoning ability of large language models through outcome-based supervision.…
Voxtral TTS
Mistral-AI, :, Alexander H. Liu +186
We introduce Voxtral TTS, an expressive multilingual text-to-speech model that generates natural speech from as little as 3 seconds of reference audio. Voxtral TTS adopts a hybrid…
Rodent-Bench
Thomas Heap, Laurence Aitchison, Emma Cahill +1
We present Rodent-Bench, a novel benchmark designed to evaluate the ability of Multimodal Large Language Models (MLLMs) to annotate rodent behaviour footage. We evaluate state-of-t…
Voxtral Realtime
Mistral-AI, :, Alexander H. Liu +166
We introduce Voxtral Realtime, a natively streaming automatic speech recognition model that matches offline transcription quality at sub-second latency. Unlike approaches that adap…
Ministral 3
Alexander H. Liu, Kartik Khandelwal, Sandeep Subramanian +116
We introduce the Ministral 3 series, a family of parameter-efficient dense language models designed for compute and memory constrained applications, available in three model sizes:…
Learning to Skip the Middle Layers of Transformers
Tim Lawson, Laurence Aitchison
Conditional computation is a popular strategy to make Transformers more efficient. Existing methods often target individual modules (e.g., mixture-of-experts layers) or skip layers…
Massively Parallel Expectation Maximization For Approximate Posteriors
Thomas Heap, Sam Bowyer, Laurence Aitchison
Bayesian inference for hierarchical models can be very challenging. MCMC methods have difficulty scaling to large models with many observations and latent variables. While variatio…
Position: Don't Use the CLT in LLM Evals With Fewer Than a Few Hundred Datapoints
Sam Bowyer, Laurence Aitchison, Desi R. Ivanova
Rigorous statistical evaluations of large language models (LLMs), including valid error bars and significance testing, are essential for meaningful and reliable performance assessm…
Jacobian Sparse Autoencoders: Sparsify Computations, Not Just Activations
Lucy Farnik, Tim Lawson, Conor Houghton +1
Sparse autoencoders (SAEs) have been successfully used to discover sparse and human-interpretable representations of the latent activations of LLMs. However, we would ultimately li…
Automated Interpretability Metrics Do Not Distinguish Trained and Random Transformers
Thomas Heap, Tim Lawson, Lucy Farnik +1
Sparse autoencoders (SAEs) are widely used to extract sparse, interpretable latents from transformer activations. We test whether commonly used SAE quality metrics and automatic ex…
Position: Bayesian Deep Learning is Needed in the Age of Large-Scale AI
Theodore Papamarkou, Maria Skoularidou, Konstantina Palla +22
In the current landscape of deep learning research, there is a predominant emphasis on achieving high predictive accuracy in supervised tasks involving large image and language dat…
Using Autodiff to Estimate Posterior Moments, Marginals and Samples
Sam Bowyer, Thomas Heap, Laurence Aitchison
Importance sampling is a popular technique in Bayesian inference: by reweighting samples drawn from a proposal distribution we are able to obtain samples and moment estimates from…