Research

Research papers, whitepapers, case studies, etc. 1271 public listings.

• Findings of the WMT25 Multilingual Instruction Shared Task: Persistent Hurdles in Reasoning (Cohere)
  The WMT25 Multilingual Instruction Shared Task (MIST) introduces a benchmark to evaluate large language models (LLMs) across 30 languages.
• Measuring the Economic Value of Open Source (Linuxfoundation)
  LINUX FOUNDATION RESEARCH We’re investigating the impact of open source collaboration to solve the world’s most pressing challenges.
• Which Prompts Make The Difference? Data Prioritization For Efficient Human LLM Evaluation (Cohere)
  Data Prioritization For Efficient Human LLM Evaluation Human evaluation of LLMs is critical, but comes at a high cost.
• LLM2Vec-Gen: Generative Embeddings from Large Language Models (Cohere)
  LLM2 Vec-Gen: Generative Embeddings from Large Language Models Fine-tuning LLM-based text embedders via contrastive learning maps inputs an…
• Command A: An Enterprise-Ready Large Language Model (Cohere)
  In this report we describe the development of Command A, a powerful large language model purpose-built to excel at real-world enterprise us…
• On the Fairness Impacts of Hardware Selection in Machine Learning (Cohere)
  In the machine learning ecosystem, hardware selection is often regarded as a mere utility, overshadowed by the spotlight on algorithms and…
• Studying the Impact of Magnitude Pruning on Contrastive Learning Methods (Cohere)
  We study the impact of different pruning techniques on the representation learned by deep neural networks trained with contrastive loss fun…
• Aya Dataset: An Open-Access Collection for Multilingual Instruction Tuning (Cohere)
  Datasets are foundational to many breakthroughs in modern artificial intelligence.
• Mitigating Harm in Language Models with Conditional-Likelihood Filtration (Cohere)
  Nov 28, 2021 Mitigating Harm in Language Models with Conditional-Likelihood Filtration Authors Helen Ngo, Cooper Raterink, João G.M.
• Commit0: Library Generation from Scratch (Cohere)
  With the goal of benchmarking generative systems beyond expert software development ability, we introduce Commit0, a benchmark that challen…
• Unlocking Reasoning Capability on Machine Translation in Large Language Models (Cohere)
  Reasoning-oriented large language models (RLMs) achieve strong gains on tasks such as mathematics and coding by generating explicit interme…
• Reverse Engineering Human Preferences with Reinforcement Learning (Cohere)
  The capabilities of Large Language Models (LLMs) are routinely evaluated by other LLMs trained to predict human preferences.
• M-RewardBench: Evaluating Reward Models in Multilingual Settings (Cohere)
  Reward models (RMs) have driven the state-of-the-art performance of LLMs today by enabling the integration of human feedback into the langu…
• MotherDuck: DuckDB in the Cloud and in the Client - MotherDuck Research Papers (Motherduck)
  MotherDuck is the serverless data warehouse built on DuckDB. Fast SQL analytics without managing infrastructure.
• Diversity, Equity, and Inclusion in Open Source (Linuxfoundation)
  LINUX FOUNDATION RESEARCH We’re investigating the impact of open source collaboration to solve the world’s most pressing challenges.
• 2026 State of Tech Talent Japan Report (Linuxfoundation)
  LINUX FOUNDATION RESEARCH We’re investigating the impact of open source collaboration to solve the world’s most pressing challenges.
• Open Source for Sustainability (Linuxfoundation)
  LINUX FOUNDATION RESEARCH We’re investigating the impact of open source collaboration to solve the world’s most pressing challenges.
• Recommended Practices for Hosting and Managing Open Source Projects on GitHub (Linuxfoundation)
  LINUX FOUNDATION RESEARCH We’re investigating the impact of open source collaboration to solve the world’s most pressing challenges.
• Constitutional Classifiers: Defending against universal jailbreaks (Anthropic)
  A paper from Anthropic describing a new way to guard LLMs against jailbreaking.
• Auditing language models for hidden objectives (Anthropic)
  A new paper from the Anthropic Alignment Science and Interpretability teams studies alignment audits—systematic investigations into whether…
• Enabling independent research on how people use Claude (Anthropic)
  Earlier this year, we ran a pilot giving external researchers access to aggregate, real-world Claude usage data.
• Project Swap: What happens when agents trade for us? (Anthropic)
  To see what works and what breaks when agents are sent into a market, we made a miniature market of Claudes.
• Forecasting rare language model behaviors (Anthropic)
  One of the major goals of Alignment Science is to predict AI models’ propensity for dangerous behaviors before those behaviors occur.
• Disempowerment patterns in real-world AI usage (Anthropic)
  Analysis of 1.5 million Claude conversations finds how often AI undermines users' autonomous judgment on beliefs, values, or actions.
• Yes, Claude can do Nine Loops (Anthropic)
  In this guest post, physicist and science writer Matt von Hippel shares what happened when he issued a challenge to AI companies regarding…
• Tracing model outputs to the training data (Anthropic)
  Scaling influence functions to 52B-parameter models shows generalization becomes more abstract and cross-lingual as models grow.
• The State of Commercial Open Source 2025 (Linuxfoundation)
  LINUX FOUNDATION RESEARCH We’re investigating the impact of open source collaboration to solve the world’s most pressing challenges.
• Measuring faithfulness in Chain-of-Thought reasoning (Anthropic)
  Does chain-of-thought reasoning reflect a model's actual decision process? We find faithfulness varies by task and declines in larger model…
• Challenges in evaluating AI systems (Anthropic)
  Six challenges of AI evaluation, from benchmark fragility to expert red teaming, and policy recommendations for better evaluation science.
• Discovering cryptographic weaknesses with Claude (Anthropic)
  Using Claude Mythos Preview, researchers at Anthropic have discovered improved ways to attack cryptographic algorithms (the mathematical me…
• An alignment assessment of recent cybersecurity incidents (Anthropic)
  We present an alignment assessment of four incidents in which Claude models gained unauthorized access to real third-party systems.
• Decomposing language models into components (Anthropic)
  Interpretability research decomposing a transformer layer into thousands of interpretable features that can steer model behavior.
• Patterns and problems in multiagent systems (Anthropic)
  We ran experiments on swarms of Claude agents and found coordination failures, collusion, and sabotage.
• Reviewing the evidence on worker retraining programs (Anthropic)
  We're sharing a review of the evidence on worker retraining programs, coauthored by independent researcher David Roodman and Anthropic's Ma…
• Sycophancy to subterfuge: Investigating reward tampering in language models (Anthropic)
  Empirical evidence that serious misalignment can emerge from seemingly benign reward misspecification.
• Open Source Innovation as a Potential Lever for Economic Recovery (Linuxfoundation)
  LINUX FOUNDATION RESEARCH We’re investigating the impact of open source collaboration to solve the world’s most pressing challenges.
• From shortcuts to sabotage: natural emergent misalignment from reward hacking (Anthropic)
  We show for the first time that realistic AI training processes can accidentally produce misaligned models.
• Making Claude a chemist (Anthropic)
  When working with molecules, chemists move between hand-drawn structures on a whiteboard, instrument readouts, database query strings, and…
• Specific versus general principles for Constitutional AI (Anthropic)
  Models trained with Constitutional AI can generalize from a single broad principle to produce harmless assistants.
• Economic Index: AI's role in the US and global economy (Anthropic)
  New research from the Anthropic Economic Index exploring geographic patterns of AI use across the US and the global economy.
• GLM-5.3 and the spread of advanced cyber capabilities (Anthropic)
  GLM-5.3 can autonomously build end-to-end cyber exploits, but unlike other frontier models, it was released without meaningful safeguards t…
• Discovering behaviors with model-written evaluations (Anthropic)
  154 automatically generated evaluation datasets reveal sycophancy and other concerning behaviors that grow with model size and RLHF.
• Measuring LLMs’ ability to develop exploits (Anthropic)
  Two new academic benchmarks measure AI models' ability to develop exploits, plus an updated benchmark for smart contract exploitation.
• Formalizing Fermat's Last Theorem (Anthropic)
  We are sharing the first complete computer-checked proof of Fermat’s Last Theorem. Claude worked largely autonomously over 11 days to write…
• Using dictionary learning features as classifiers (Anthropic)
  At the link above, we report some developing work from the Anthropic interpretability team on developing feature-based classifiers, which m…
• 2025 State of Tech Talent Report (Linuxfoundation)
  LINUX FOUNDATION RESEARCH We’re investigating the impact of open source collaboration to solve the world’s most pressing challenges.
• The Value of Open Source AI for the Canadian Economy (Linuxfoundation)
  LINUX FOUNDATION RESEARCH We’re investigating the impact of open source collaboration to solve the world’s most pressing challenges.
• LLMs with cyber toolkits can conduct multistage cyber operations on business-sized computer networks (Anthropic)
  Large Language Models (LLMs) that are not fine-tuned for cybersecurity can succeed in multistage attacks on networks with dozens of hosts w…
• Towards understanding sycophancy in language models (Anthropic)
  Five state-of-the-art AI assistants consistently exhibit sycophancy, suggesting human feedback training partly drives the behavior.
• Next-generation Constitutional Classifiers (Anthropic)
  Over time, we’ve implemented a variety of protections that have made our models much less likely to assist with dangerous user queries—in p…