Today’s most advanced vision-language models (VLMs) remain proprietary. The strongest open-weight models rely heavily on synthetic data from proprietary VLMs to achieve good performance, effectively distilling these closed VLMs into open ones. As a result, the community has been missing foundational knowledge about how to build performant VLMs from scratch. We present Molmo, a new family of VLMs that are state-of-the-art in their class of openness. Our key contribution is a collection of new datasets called PixMo, including a dataset of highly detailed image captions for pre-training, a free-form image Q&A dataset for fine-tuning, and an innovative 2D pointing dataset, all collected without the use of external VLMs. The success of our approach relies on careful modeling choices, a well- tuned training pipeline, and, most critically, the quality of our newly collected datasets. Our best-in-class 72B model not only outperforms others in the class of open weight and data models, but also outperforms larger proprietary models including Claude 3.5 Sonnet, and Gemini 1.5 Pro and Flash, second only to GPT-4o based on both academic benchmarks and on a large human evaluation. Our model weights, new datasets, and source code are available at https://molmo.allenai.org/blog.

Niklas Muennighoff, Zitong Yang, Weijia Shi, Xiang Lisa Li, Li Fei-Fei, Hannaneh Hajishirzi, Luke Zettlemoyer, Percy Liang, Emmanuel Candes, Tatsunori Hashimoto. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. 2025.

Image representations are often evaluated through disjointed, task-specific protocols, leading to a fragmented understanding of model capabilities. For instance, it is unclear whether an image embedding model adept at clustering images is equally good at retrieving relevant images given a piece of text. We introduce the Massive Image Embedding Benchmark (MIEB) to evaluate the performance of image and image-text embedding models across the broadest spectrum to date. MIEB spans 38 languages across 130 individual tasks, which we group into 8 high-level categories. We benchmark 50 models across our benchmark, finding that no single method dominates across all task categories. We reveal hidden capabilities in advanced vision models such as their accurate visual representation of texts, and their yet limited capabilities in interleaved encodings and matching images and texts in the presence of confounders. We also show that the performance of vision encoders on MIEB correlates highly with their performance when used in multimodal large language models. Our code, dataset, and leaderboard are publicly available at https://github.com/embeddings-benchmark/mteb.

Software is one of the most powerful tools that we humans have at our disposal; it allows a skilled programmer to interact with the world in complex and profound ways. At the same time, thanks to improvements in large language models (LLMs), there has also been a rapid development in AI agents that interact with and effect change in their surrounding environments. In this paper, we introduce OpenHands, a platform for the development of powerful and flexible AI agents that interact with the world in similar ways to a human developer: by writing code, interacting with a command line, and browsing the web. We describe how the platform allows for the implementation of new agents, utilization of various LLMs, safe interaction with sandboxed environments for code execution, and incorporation of evaluation benchmarks. Based on our currently incorporated benchmarks, we perform an evaluation of agents over 13 challenging tasks, including software engineering (e.g., SWE-Bench) and web browsing (e.g., WebArena), amongst others. Released under the permissive MIT license, OpenHands is a community project spanning academia and industry with more than 2K contributions from over 186 contributors in less than six months of development, and will improve going forward.

Low precision training and inference affect both the quality and cost of language models, but current scaling laws do not account for this. In this work, we devise "precision-aware" scaling laws for both training and inference. We propose that training in lower precision reduces the model's "effective parameter count," allowing us to predict the additional loss incurred from training in low precision and post-train quantization. For inference, we find that the degradation introduced by post-training quantization increases as models are trained on more data, eventually making additional pretraining data actively harmful. For training, our scaling laws allow us to predict the loss of a model with different parts in different precisions, and suggest that training larger models in lower precision can be compute optimal. We unify the scaling laws for post and pretraining quantization to arrive at a single functional form that predicts degradation from training and inference in varied precisions. We fit on over 465 pretraining runs and validate our predictions on model sizes up to 1.7B parameters trained on up to 26B tokens.

The data mixture for large language model pre-training significantly impacts performance, yet how to determine an effective mixture remains unclear. We propose RegMix to automatically identify a high-performing data mixture by formulating it as a regression task. RegMix trains many small models on diverse data mixtures, uses regression to predict performance of unseen mixtures, and applies the best predicted mixture to train a large-scale model with orders of magnitude more compute. To empirically validate RegMix, we train 512 models with 1M parameters for 1B tokens to fit the regression model and predict the best data mixture. Using this mixture we train a 1B parameter model for 25B tokens (i.e. 1000× larger and 25× longer) which we find performs best among 64 candidate 1B parameter models with other mixtures. Furthermore, RegMix consistently outperforms human selection in experiments involving models up to 7B models trained on 100B tokens, while matching or exceeding DoReMi using just 10% of the computational resources. Our experiments also show that (1) Data mixtures significantly impact performance; (2) Web corpora rather than data perceived as high-quality like Wikipedia have the strongest positive correlation with downstream performance; (3) Domains interact in complex ways often contradicting common sense, thus automatic approaches like RegMix are needed; (4) Data mixture effects transcend scaling laws. Our code is available at https://github.com/sail-sg/regmix.

All text-based language problems can be reduced to either generation or embedding. Current models only perform well at one or the other. We introduce generative representational instruction tuning (GRIT) whereby a large language model is trained to handle both generative and embedding tasks by distinguishing between them through instructions. Compared to other open models, our resulting GritLM-7B is among the top models on the Massive Text Embedding Benchmark (MTEB) and outperforms various models up to its size on a range of generative tasks. By scaling up further, GritLM-8x7B achieves even stronger generative performance while still being among the best embedding models. Notably, we find that GRIT matches training on only generative or embedding data, thus we can unify both at no performance loss. Among other benefits, the unification via GRIT speeds up Retrieval-Augmented Generation (RAG) by > 60% for long documents, by no longer requiring separate retrieval and generation models. Models, code, etc. are freely available at https://github.com/ContextualAI/gritlm.

We introduce OLMoE, a fully open, state-of-the-art language model leveraging sparse Mixture-of-Experts (MoE). OLMoE-1B-7B has 7 billion (B) parameters but uses only 1B per input token. We pretrain it on 5 trillion tokens and further adapt it to create OLMoE-1B-7B-Instruct. Our models outperform all available models with similar active parameters, even surpassing larger ones like Llama2-13B-Chat and DeepSeekMoE-16B. We present novel findings on MoE training, define and analyze new routing properties showing high specialization in our model, and open-source all our work: model weights, training data, code, and logs.

Existing retrieval benchmarks primarily consist of information-seeking queries (e.g., aggregated questions from search engines) where keyword or semantic-based retrieval is usually sufficient. However, many complex real-world queries require in-depth reasoning to identify relevant documents that go beyond surface form matching. For example, finding documentation for a coding question requires understanding the logic and syntax of the functions involved. To better benchmark retrieval on such challenging queries, we introduce BRIGHT, the first text retrieval benchmark that requires intensive reasoning to retrieve relevant documents. Our dataset consists of 1,398 real-world queries spanning diverse domains such as economics, psychology, mathematics, coding, and more. These queries are drawn from naturally occurring or carefully curated human data. Extensive evaluation reveals that even state-of-the-art retrieval models perform poorly on BRIGHT. The leading model on the MTEB leaderboard (Muennighoff et al., 2023), which achieves a score of 59.0 nDCG@10,1 produces a score of nDCG@10 of 18.0 on BRIGHT. We show that incorporating explicit reasoning about the query improves retrieval performance by up to 12.2 points. Moreover, incorporating retrieved documents from the top-performing retriever boosts question answering performance by over 6.6 points. We believe that BRIGHT paves the way for future research on retrieval systems in more realistic and challenging settings.

Autonomous systems for software engineering are now capable of fixing bugs and developing features. These systems are commonly evaluated on SWE-bench (Jimenez et al., 2024a), which assesses their ability to solve software issues from GitHub repositories. However, SWE-bench uses only Python repositories, with problem statements presented predominantly as text and lacking visual elements such as images. This limited coverage motivates our inquiry into how existing systems might perform on unrepresented software engineering domains (e.g., front-end, game development, DevOps), which use different programming languages and paradigms. Therefore, we propose SWE-bench Multimodal (SWE-bench M), to evaluate systems on their ability to fix bugs in visual, user-facing JavaScript software. SWE-bench M features 617 task instances collected from 17 JavaScript libraries used for web interface design, diagramming, data visualization, syntax highlighting, and interactive mapping. Each SWE-bench M task instance contains at least one image in its problem statement or unit tests. Our analysis finds that top-performing SWE-bench systems struggle with SWE-bench M, revealing limitations in visual problem-solving and cross-language generalization. Lastly, we show that SWE-agent’s flexible language-agnostic features enable it to substantially outperform alternatives on SWE-bench M, resolving 12% of task instances compared to 6% for the next best system.

Guijin Son, Hanwool Lee, Sungdong Kim, Seungone Kim, Niklas Muennighoff, Taekyoon Choi, Cheonbok Park, Kang Min Yoo, Stella Biderman. Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2025.

We introduce FlexOLMo, a new class of language models (LMs) that supports (1) distributed training without data sharing, where different model parameters are independently trained on private datasets, and (2) data-flexible inference, where these parameters along with their associated data can be easily included or excluded from model inferences with no further training. FlexOLMo employs a mixture-of-experts (MoE) architecture where each expert is trained independently on private datasets and later integrated through a new nonparametric routing without any joint training across datasets. FlexOLMo is trained on FLEXMIX, a corpus we curate comprising seven restricted sets, either real or realistic approximations, alongside publicly available datasets. We evaluate models with up to 37 billion parameters (20 billion active) on 31 diverse downstream tasks. We show that a general expert trained on public data can be effectively combined with independently trained experts from other data owners significantly benefiting from these restricted sets (an average 41% relative improvement) while allowing flexible opt-out at inference time (e.g., for users without appropriate licenses or permissions). Our approach also outperforms prior model merging methods by 10.1% on average and surpasses the standard MoE trained without data restrictions using the same training FLOPs. Altogether, FlexOLMo enables training on restricted data while keeping data local and supports fine-grained control of data access at inference.

The current trend of scaling language models involves increasing both parameter count and training data set size. Extrapolating this trend suggests that training data set size may soon be limited by the amount of text data available on the internet. Motivated by this limit, we investigate scaling language models in data-constrained regimes. Specifically, we run a large set of experiments varying the extent of data repetition and compute budget, ranging up to 900 billion training tokens and 9 billion parameter models. We find that with constrained data for a fixed compute budget, training with up to 4 epochs of repeated data yields negligible changes to loss compared to having unique data. However, with more repetition, the value of adding compute eventually decays to zero. We propose and empirically validate a scaling law for compute optimality that accounts for the decreasing value of repeated tokens and excess parameters. Finally, we experiment with approaches mitigating data scarcity, including augmenting the training data set with code data or removing commonly used filters. Models and data sets from our 400 training runs are freely available at https://github.com/huggingface/datablations.

Dirk Groeneveld, Iz Beltagy, Evan Walsh, Akshita Bhagia, Rodney Kinney, Oyvind Tafjord, Ananya Jha, Hamish Ivison, Ian Magnusson, Yizhong Wang, Shane Arora, David Atkinson, Russell Authur, Khyathi Chandu, Arman Cohan, Jennifer Dumas, Yanai Elazar, Yuling Gu, Jack Hessel, Tushar Khot, William Merrill, Jacob Morrison, Niklas Muennighoff, Aakanksha Naik, Crystal Nam, Matthew Peters, Valentina Pyatkin, Abhilasha Ravichander, Dustin Schwenk, Saurabh Shah, William Smith, Emma Strubell, Nishant Subramani, Mitchell Wortsman, Pradeep Dasigi, Nathan Lambert, Kyle Richardson, Luke Zettlemoyer, Jesse Dodge, Kyle Lo, Luca Soldaini, Noah Smith, Hannaneh Hajishirzi. Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2024.

Shivalika Singh, Freddie Vargus, Daniel D’souza, Börje F. Karlsson, Abinaya Mahendiran, Wei-Yin Ko, Herumb Shandilya, Jay Patel, Deividas Mataciunas, Laura O’Mahony, Mike Zhang, Ramith Hettiarachchi, Joseph Wilson, Marina Machado, Luisa Moura, Dominik Krzemiński, Hakimeh Fadaei, Irem Ergun, Ifeoma Okoh, Aisha Alaagib, Oshan Mudannayake, Zaid Alyafeai, Vu Chien, Sebastian Ruder, Surya Guthikonda, Emad Alghamdi, Sebastian Gehrmann, Niklas Muennighoff, Max Bartolo, Julia Kreutzer, Ahmet Üstün, Marzieh Fadaee, Sara Hooker. Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2024.

Luca Soldaini, Rodney Kinney, Akshita Bhagia, Dustin Schwenk, David Atkinson, Russell Authur, Ben Bogin, Khyathi Chandu, Jennifer Dumas, Yanai Elazar, Valentin Hofmann, Ananya Jha, Sachin Kumar, Li Lucy, Xinxi Lyu, Nathan Lambert, Ian Magnusson, Jacob Morrison, Niklas Muennighoff, Aakanksha Naik, Crystal Nam, Matthew Peters, Abhilasha Ravichander, Kyle Richardson, Zejiang Shen, Emma Strubell, Nishant Subramani, Oyvind Tafjord, Evan Walsh, Luke Zettlemoyer, Noah Smith, Hannaneh Hajishirzi, Iz Beltagy, Dirk Groeneveld, Jesse Dodge, Kyle Lo. Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2024.

Ahmet Üstün, Viraat Aryabumi, Zheng Yong, Wei-Yin Ko, Daniel D’souza, Gbemileke Onilude, Neel Bhandari, Shivalika Singh, Hui-Lee Ooi, Amr Kayid, Freddie Vargus, Phil Blunsom, Shayne Longpre, Niklas Muennighoff, Marzieh Fadaee, Julia Kreutzer, Sara Hooker. Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2024.

Holy Lovenia, Rahmad Mahendra, Salsabil Maulana Akbar, Lester James V. Miranda, Jennifer Santoso, Elyanah Aco, Akhdan Fadhilah, Jonibek Mansurov, Joseph Marvin Imperial, Onno P. Kampman, Joel Ruben Antony Moniz, Muhammad Ravi Shulthan Habibi, Frederikus Hudi, Railey Montalan, Ryan Ignatius, Joanito Agili Lopo, William Nixon, Börje F. Karlsson, James Jaya, Ryandito Diandaru, Yuze Gao, Patrick Amadeus, Bin Wang, Jan Christian Blaise Cruz, Chenxi Whitehouse, Ivan Halim Parmonangan, Maria Khelli, Wenyu Zhang, Lucky Susanto, Reynard Adha Ryanda, Sonny Lazuardi Hermawan, Dan John Velasco, Muhammad Dehan Al Kautsar, Willy Fitra Hendria, Yasmin Moslem, Noah Flynn, Muhammad Farid Adilazuarda, Haochen Li, Johanes Lee, R. Damanhuri, Shuo Sun, Muhammad Reza Qorib, Amirbek Djanibekov, Wei Qi Leong, Quyet V. Do, Niklas Muennighoff, Tanrada Pansuwan, Ilham Firdausi Putra, Yan Xu, Tai Ngee Chia, Ayu Purwarianti, Sebastian Ruder, William Tjhi, Peerat Limkonchotiwat, Alham Fikri Aji, Sedrick Keh, Genta Indra Winata, Ruochen Zhang, Fajri Koto, Zheng-Xin Yong, Samuel Cahyawijaya. Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing. 2024.

Finetuning large language models (LLMs) on instructions leads to vast performance improvements on natural language tasks. We apply instruction tuning using code, leveraging the natural structure of Git commits, which pair code changes with human instructions. We compile CommitPack: 4 terabytes of Git commits across 350 programming languages. We benchmark CommitPack against other natural and synthetic code instructions (xP3x, Self-Instruct, OASST) on the 16B parameter StarCoder model, and achieve state-of-the-art performance among models not trained on OpenAI outputs, on the HumanEval Python benchmark (46.2% pass@1). We further introduce HumanEvalPack, expanding the HumanEval benchmark to a total of 3 coding tasks (Code Repair, Code Explanation, Code Synthesis) across 6 languages (Python, JavaScript, Java, Go, C++, Rust). Our models, OctoCoder and OctoGeeX, achieve the best performance across HumanEvalPack among all permissive models, demonstrating CommitPack's benefits in generalizing to a wider set of languages and natural coding tasks. Code, models and data are freely available at https://github.com/bigcode-project/octopack.

Kahneman & Tversky’s $\textit{prospect theory}$ tells us that humans perceive random variables in a biased but well-defined manner (1992); for example, humans are famously loss-averse. We show that objectives for aligning LLMs with human feedback implicitly incorporate many of these biases—the success of these objectives (e.g., DPO) over cross-entropy minimization can partly be ascribed to them belonging to a family of loss functions that we call $\textit{human-aware losses}$ (HALOs). However, the utility functions these methods attribute to humans still differ from those in the prospect theory literature. Using a Kahneman-Tversky model of human utility, we propose a HALO that directly maximizes the utility of generations instead of maximizing the log-likelihood of preferences, as current methods do. We call this approach KTO, and it matches or exceeds the performance of preference-based methods at scales from 1B to 30B, despite only learning from a binary signal of whether an output is desirable. More broadly, our work suggests that there is no one HALO that is universally superior; the best loss depends on the inductive biases most appropriate for a given setting, an oft-overlooked consideration.

The evaluation of English text embeddings has transitioned from evaluating a handful of datasets to broad coverage across many tasks through benchmarks such as MTEB. However, this is not the case for multilingual text embeddings due to a lack of available benchmarks. To address this problem, we introduce the Scandinavian Embedding Benchmark (SEB). SEB is a comprehensive framework that enables text embedding evaluation for Scandinavian languages across 24 tasks, 10 subtasks, and 4 task categories. Building on SEB, we evaluate more than 26 models, uncovering significant performance disparities between public and commercial solutions not previously captured by MTEB. We open-source SEB and integrate it with MTEB, thus bridging the text embedding evaluation gap for Scandinavian languages.

We introduce DataComp for Language Models, a testbed for controlled dataset experiments with the goal of improving language models.As part of DCLM, we provide a standardized corpus of 240T tokens extracted from Common Crawl, effective pretraining recipes based on the OpenLM framework, and a broad suite of 53 downstream evaluations.Participants in the DCLM benchmark can experiment with data curation strategies such as deduplication, filtering, and data mixing atmodel scales ranging from 412M to 7B parameters.As a baseline for DCLM, we conduct extensive experiments and find that model-based filtering is key to assembling a high-quality training set.The resulting dataset, DCLM-Baseline, enables training a 7B parameter language model from scratch to 63% 5-shot accuracy on MMLU with 2T training tokens.Compared to MAP-Neo, the previous state-of-the-art in open-data language models, DCLM-Baseline represents a 6 percentage point improvement on MMLU while being trained with half the compute.Our results highlight the importance of dataset design for training language models and offer a starting point for further research on data curation. We release the \dclm benchmark, framework, models, and datasets at https://www.datacomp.ai/dclm/

General-purpose artificial intelligence (AI) systems are built on massive swathes of public web data, assembled into corpora such as C4, RefinedWeb, and Dolma. To our knowledge, we conduct the first, large-scale, longitudinal audit of the consent protocols for the web domains underlying AI training corpora. Our audit of 14,000 web domains provides an expansive view of crawlable web data and how codified data use preferences are changing over time. We observe a proliferation of AI-specific clauses to limit use, acute differences in restrictions on AI developers, as well as general inconsistencies between websites' expressed intentions in their Terms of Service and their robots.txt. We diagnose these as symptoms of ineffective web protocols, not designed to cope with the widespread re-purposing of the internet for AI. Our longitudinal analyses show that in a single year (2023-2024) there has been a rapid crescendo of data restrictions from web sources, rendering ~5\%+ of all tokens in C4, or 28%+ of the most actively maintained, critical sources in C4, fully restricted from use. For Terms of Service crawling restrictions, a full 45% of C4 is now restricted. If respected or enforced, these restrictions are rapidly biasing the diversity, freshness, and scaling laws for general-purpose AI systems. We hope to illustrate the emerging crises in data consent, for both developers and creators. The foreclosure of much of the open web will impact not only commercial AI, but also non-commercial AI and academic research.

Research on scaling large language models (LLMs) has primarily focused on model parameters and training data size, overlooking the role of vocabulary size. We investigate how vocabulary size impacts LLM scaling laws by training models ranging from 33M to 3B parameters on up to 500B characters with various vocabulary configurations. We propose three complementary approaches for predicting the compute-optimal vocabulary size: IsoFLOPs analysis, derivative estimation, and parametric fit of the loss function. Our approaches converge on the conclusion that the optimal vocabulary size depends on the compute budget, with larger models requiring larger vocabularies. Most LLMs, however, use insufficient vocabulary sizes. For example, we predict that the optimal vocabulary size of Llama2-70B should have been at least 216K, 7 times larger than its vocabulary of 32K. We validate our predictions empirically by training models with 3B parameters across different FLOPs budgets. Adopting our predicted optimal vocabulary size consistently improves downstream performance over commonly used vocabulary sizes. By increasing the vocabulary size from the conventional 32K to 43K, we improve performance on ARC-Challenge from 29.1 to 32.0 with the same 2.3e21 FLOPs. Our work highlights the importance of jointly considering tokenization and model scaling for efficient pre-training. The code and demo are available at https://github.com/sail-sg/scaling-with-vocab and https://hf.co/spaces/sail/scaling-with-vocab-demo.

We introduce C-Pack, a package of resources that significantly advances the field of general text embeddings for Chinese. C-Pack includes three critical resources. 1) C-MTP is a massive training dataset for text embedding, which is based on the curation of vast unlabeled corpora and the integration of high-quality labeled corpora. 2) C-MTEB is a comprehensive benchmark for Chinese text embeddings covering 6 tasks and 35 datasets. 3) BGE is a family of embedding models covering multiple sizes. Our models outperform all prior Chinese text embeddings on C-MTEB by more than +10% upon the time of the release. We also integrate and optimize the entire suite of training methods for BGE. Along with our resources on general Chinese embedding, we release our data and models for English text embeddings. The English models also achieve state-of-the-art performance on the MTEB benchmark; meanwhile, our released English data is 2 times larger than the Chinese data. Both Chinese and English datasets are the largest public release of training data for text embeddings. All these resources are made publicly available at https://github.com/FlagOpen/FlagEmbedding.

Niklas Muennighoff, Thomas Wang, Lintang Sutawika, Adam Roberts, Stella Biderman, Teven Le Scao, M Saiful Bari, Sheng Shen, Zheng Xin Yong, Hailey Schoelkopf, Xiangru Tang, Dragomir Radev, Alham Fikri Aji, Khalid Almubarak, Samuel Albanie, Zaid Alyafeai, Albert Webson, Edward Raff, Colin Raffel. Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2023.

Zheng Xin Yong, Hailey Schoelkopf, Niklas Muennighoff, Alham Fikri Aji, David Ifeoluwa Adelani, Khalid Almubarak, M Saiful Bari, Lintang Sutawika, Jungo Kasai, Ahmed Baruwa, Genta Winata, Stella Biderman, Edward Raff, Dragomir Radev, Vassilina Nikoulina. Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2023.

Text embeddings are commonly evaluated on a small set of datasets from a single task not covering their possible applications to other tasks.It is unclear whether state-of-the-art embeddings on semantic textual similarity (STS) can be equally well applied to other tasks like clustering or reranking.This makes progress in the field difficult to track, as various models are constantly being proposed without proper evaluation.To solve this problem, we introduce the Massive Text Embedding Benchmark (MTEB).MTEB spans 8 embedding tasks covering a total of 58 datasets and 112 languages.Through the benchmarking of 33 models on MTEB, we establish the most comprehensive benchmark of text embeddings to date.We find that no particular text embedding method dominates across all tasks.This suggests that the field has yet to converge on a universal text embedding method and scale it up sufficiently to provide state-of-the-art results on all embedding tasks.MTEB comes with open-source code and a public leaderboard at https://github.com/ embeddings-benchmark/mteb.

Risto Luukkonen, Ville Komulainen, Jouni Luoma, Anni Eskelinen, Jenna Kanerva, Hanna-Mari Kupari, Filip Ginter, Veronika Laippala, Niklas Muennighoff, Aleksandra Piktus, Thomas Wang, Nouamane Tazi, Teven Scao, Thomas Wolf, Osma Suominen, Samuli Sairanen, Mikko Merioksa, Jyrki Heinonen, Aija Vahtola, Samuel Antao, Sampo Pyysalo. Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing. 2023.

The current trend of scaling language models involves increasing both parameter count and training dataset size. Extrapolating this trend suggests that training dataset size may soon be limited by the amount of text data available on the internet. Motivated by this limit, we investigate scaling language models in data-constrained regimes. Specifically, we run a large set of experiments varying the extent of data repetition and compute budget, ranging up to 900 billion training tokens and 9 billion parameter models. We find that with constrained data for a fixed compute budget, training with up to 4 epochs of repeated data yields negligible changes to loss compared to having unique data. However, with more repetition, the value of adding compute eventually decays to zero. We propose and empirically validate a scaling law for compute optimality that accounts for the decreasing value of repeated tokens and excess parameters. Finally, we experiment with approaches mitigating data scarcity, including augmenting the training dataset with code data or removing commonly used filters. Models and datasets from our 400 training runs are freely available at https://github.com/huggingface/datablations.