25 citations · 54 across the 19 of their papers we have counts for
19 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…
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…
A Family of LLMs Liberated from Static Vocabularies
Aleph Alpha, :, Adnen Abdessaied +35
Tokenization is a central component of natural language processing in current large language models (LLMs), enabling models to convert raw text into processable units. Although lea…
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…
The Impact of Inference Acceleration on Bias of LLMs
Elisabeth Kirsten, Ivan Habernal, Vedant Nanda +1
Last few years have seen unprecedented advances in capabilities of Large Language Models (LLMs). These advancements promise to benefit a vast array of application domains. However,…
Understanding Memorisation in LLMs: Dynamics, Influencing Factors, and Implications
Till Speicher, Mohammad Aflah Khan, Qinyuan Wu +5
Understanding whether and to what extent large language models (LLMs) have memorised training data has important implications for the reliability of their output and the privacy of…
Understanding the Role of Invariance in Transfer Learning
Till Speicher, Vedant Nanda, Krishna P. Gummadi
Transfer learning is a powerful technique for knowledge-sharing between different tasks. Recent work has found that the representations of models with certain invariances, such as…
Lawma: The Power of Specialization for Legal Annotation
Ricardo Dominguez-Olmedo, Vedant Nanda, Rediet Abebe +6
Annotation and classification of legal text are central components of empirical legal research. Traditionally, these tasks are often delegated to trained research assistants. Motiv…
Towards Reliable Latent Knowledge Estimation in LLMs: Zero-Prompt Many-Shot Based Factual Knowledge Extraction
Qinyuan Wu, Mohammad Aflah Khan, Soumi Das +7
In this paper, we focus on the challenging task of reliably estimating factual knowledge that is embedded inside large language models (LLMs). To avoid reliability concerns with pr…
What Happens During Finetuning of Vision Transformers: An Invariance Based Investigation
Gabriele Merlin, Vedant Nanda, Ruchit Rawal +1
The pretrain-finetune paradigm usually improves downstream performance over training a model from scratch on the same task, becoming commonplace across many areas of machine learni…
Diffused Redundancy in Pre-trained Representations
Vedant Nanda, Till Speicher, John P. Dickerson +3
Representations learned by pre-training a neural network on a large dataset are increasingly used successfully to perform a variety of downstream tasks. In this work, we take a clo…
Investigating the Effects of Fairness Interventions Using Pointwise Representational Similarity
Camila Kolling, Till Speicher, Vedant Nanda +2
Machine learning (ML) algorithms can often exhibit discriminatory behavior, negatively affecting certain populations across protected groups. To address this, numerous debiasing me…
Measuring Representational Robustness of Neural Networks Through Shared Invariances
Vedant Nanda, Till Speicher, Camila Kolling +3
A major challenge in studying robustness in deep learning is defining the set of ``meaningless'' perturbations to which a given Neural Network (NN) should be invariant. Most work o…
Rawlsian Fairness in Online Bipartite Matching: Two-sided, Group, and Individual
Seyed A. Esmaeili, Sharmila Duppala, Davidson Cheng +3
Online bipartite-matching platforms are ubiquitous and find applications in important areas such as crowdsourcing and ridesharing. In the most general form, the platform consists o…
Do Invariances in Deep Neural Networks Align with Human Perception?
Vedant Nanda, Ayan Majumdar, Camila Kolling +4
An evaluation criterion for safe and trustworthy deep learning is how well the invariances captured by representations of deep neural networks (DNNs) are shared with humans. We ide…
Unifying Model Explainability and Robustness via Machine-Checkable Concepts
Vedant Nanda, Till Speicher, John P. Dickerson +2
As deep neural networks (DNNs) get adopted in an ever-increasing number of applications, explainability has emerged as a crucial desideratum for these models. In many real-world ta…
Fairness Through Robustness: Investigating Robustness Disparity in Deep Learning
Vedant Nanda, Samuel Dooley, Sahil Singla +2
Deep neural networks (DNNs) are increasingly used in real-world applications (e.g. facial recognition). This has resulted in concerns about the fairness of decisions made by these…
Balancing the Tradeoff between Profit and Fairness in Rideshare Platforms During High-Demand Hours
Vedant Nanda, Pan Xu, Karthik Abinav Sankararaman +2
Rideshare platforms, when assigning requests to drivers, tend to maximize profit for the system and/or minimize waiting time for riders. Such platforms can exacerbate biases that d…
On the Long-term Impact of Algorithmic Decision Policies: Effort Unfairness and Feature Segregation through Social Learning
Hoda Heidari, Vedant Nanda, Krishna P. Gummadi
Most existing notions of algorithmic fairness are one-shot: they ensure some form of allocative equality at the time of decision making, but do not account for the adverse impact o…