Background image for aesthetic purposes

Research papers

Work by Cohere Labs and Technical Staff at Cohere

Learn more about our lab

Filter papers

Remove All Filters

Feb 26, 2025

When Personalization Meets Reality: A Multi-Faceted Analysis of Personalized Preference Learning

Dec 18, 2024

Bridging the Data Provenance Gap Across Text, Speech, and Video

Dec 10, 2024

Policy Primer - Translating Safety

AI Policy

multilingual

Safety

Dec 02, 2024

The Reality of AI and Biorisk

AI Policy

Responsible AI

Safety

Oct 15, 2024

Mix Data or Merge Models? Optimizing for Diverse Multi-Task Learning

multilingual

Safety

Supervised Learning

Language

Human Feedback

Efficiency

Aug 29, 2024

The Future of International Scientific Assessments of AI’s Risks

AI Policy

Safety

Jul 19, 2024

Consent in Crisis: The Rapid Decline of the AI Data Commons

Responsible AI

Safety

AI Policy

Data

Jul 09, 2024

On the Limitations of Compute Thresholds as a Governance Strategy

Jun 08, 2024

The Multilingual Alignment Prism: Aligning Global and Local Preferences to Reduce Harm

multilingual

Safety

Supervised Learning

Language

Human Feedback

Apr 29, 2024

Replacing Judges with Juries: Evaluating LLM Generations with a Panel of Diverse Models

Apr 24, 2024

The PRISM Alignment Project: What Participatory, Representative and Individualised Human Feedback Reveals About the Subjective and Multicultural Alignment of Large Language Models

Mar 07, 2024

From One to Many: Expanding the Scope of Toxicity Mitigation in Language Models

Language

Safety

Generative Models

Continual Learning

Feb 29, 2024

Here's a Free Lunch: Sanitizing Backdoored Models with Model Merge

Oct 25, 2023

The Data Provenance Initiative: A Large Scale Audit of Dataset Licensing & Attribution in AI

Responsible AI

Safety

AI Policy

Data

Oct 24, 2023

Locally Differentially Private Document Generation Using Zero Shot Prompting

Safety

Privacy

Oct 11, 2023

Goodtriever: Adaptive Toxicity Mitigation with Retrieval-augmented Models

Safety

Generative Models

Language Models

Aug 31, 2023

Sparkles: Unlocking Chats Across Multiple Images for Multimodal Instruction-Following Models

Jun 14, 2023

The Presidio Recommendations on Responsible Generative AI - World Economic Forum

Interpretability

Safety

Responsible AI

AI Policy

May 07, 2023

Language Models Don't Always Say What They Think: Unfaithful Explanations in Chain-of-Thought Prompting

Safety

Reasoning

Interpretability

Apr 24, 2023

On the Challenges of Using Black-Box APIs for Toxicity Evaluation in Research

Safety

Reproducibility

Responsible AI