Correlating and Predicting Human Evaluations of Language Models from Natural Language Processing Benchmarks

The explosion of high-performing conversational language models (LMs) has spurred a shift from classic natural language processing (NLP) benchmarks to expensive, time-consuming and noisy human evaluations - yet the relationship between these two evaluation strategies remains hazy. In this paper, we conduct a large-scale study of four Chat Llama 2 models, comparing their performance on 160 standard NLP benchmarks (e.g., MMLU, ARC, BIG-Bench Hard) against extensive human preferences on more than 11k single-turn and 2k multi-turn dialogues from over 2k human annotators. Our findings are striking: most NLP benchmarks strongly correlate with human evaluations, suggesting that cheaper, automated metrics can serve as surprisingly reliable predictors of human preferences. Three human evaluations, such as adversarial dishonesty and safety, are anticorrelated with NLP benchmarks, while two are uncorrelated. Moreover, through overparameterized linear regressions, we show that NLP scores can accurately predict human evaluations across different model scales, offering a path to reduce costly human annotation without sacrificing rigor. Overall, our results affirm the continued value of classic benchmarks and illuminate how to harness them to anticipate real-world user satisfaction - pointing to how NLP benchmarks can be leveraged to meet evaluation needs of our new era of conversational AI.

Law of the Weakest Link: Cross Capabilities of Large Language Models

The development and evaluation of Large Language Models (LLMs) have largely focused on individual capabilities. However, this overlooks the intersection of multiple abilities across different types of expertise that are often required for real-world tasks, which we term cross capabilities. To systematically explore this concept, we first define seven core individual capabilities and then pair them to form seven common cross capabilities, each supported by a manually constructed taxonomy. Building on these definitions, we introduce CrossEval, a benchmark comprising 1,400 human-annotated prompts, with 100 prompts for each individual and cross capability. To ensure reliable evaluation, we involve expert annotators to assess 4,200 model responses, gathering 8,400 human ratings with detailed explanations to serve as reference examples. Our findings reveal that, in both static evaluations and attempts to enhance specific abilities, current LLMs consistently exhibit the "Law of the Weakest Link," where cross-capability performance is significantly constrained by the weakest component. Specifically, across 58 cross-capability scores from 17 models, 38 scores are lower than all individual capabilities, while 20 fall between strong and weak, but closer to the weaker ability. These results highlight the under-performance of LLMs in cross-capability tasks, making the identification and improvement of the weakest capabilities a critical priority for future research to optimize performance in complex, multi-dimensional scenarios.

The Llama 3 Herd of Models

Modern artificial intelligence (AI) systems are powered by foundation models. This paper presents a new set of foundation models, called Llama 3. It is a herd of language models that natively support multilinguality, coding, reasoning, and tool usage. Our largest model is a dense Transformer with 405B parameters and a context window of up to 128K tokens. This paper presents an extensive empirical evaluation of Llama 3. We find that Llama 3 delivers comparable quality to leading language models such as GPT-4 on a plethora of tasks. We publicly release Llama 3, including pre-trained and post-trained versions of the 405B parameter language model and our Llama Guard 3 model for input and output safety. The paper also presents the results of experiments in which we integrate image, video, and speech capabilities into Llama 3 via a compositional approach. We observe this approach performs competitively with the state-of-the-art on image, video, and speech recognition tasks. The resulting models are not yet being broadly released as they are still under development.

Quantifying Variance in Evaluation Benchmarks

Evaluation benchmarks are the cornerstone of measuring capabilities of large language models (LLMs), as well as driving progress in said capabilities. Originally designed to make claims about capabilities (or lack thereof) in fully pretrained models, evaluation benchmarks are now also extensively used to decide between various training choices. Despite this widespread usage, we rarely quantify the variance in our evaluation benchmarks, which dictates whether differences in performance are meaningful. Here, we define and measure a range of metrics geared towards measuring variance in evaluation benchmarks, including seed variance across initialisations, and monotonicity during training. By studying a large number of models -- both openly available and pretrained from scratch -- we provide empirical estimates for a variety of variance metrics, with considerations and recommendations for practitioners. We also evaluate the utility and tradeoffs of continuous versus discrete performance measures and explore options for better understanding and reducing this variance. We find that simple changes, such as framing choice tasks (like MMLU) as completion tasks, can often reduce variance for smaller scale ($\sim$7B) models, while more involved methods inspired from human testing literature (such as item analysis and item response theory) struggle to meaningfully reduce variance. Overall, our work provides insights into variance in evaluation benchmarks, suggests LM-specific techniques to reduce variance, and more generally encourages practitioners to carefully factor in variance when comparing models.

Effective Long-Context Scaling of Foundation Models

We present a series of long-context LLMs that support effective context windows of up to 32,768 tokens. Our model series are built through continual pretraining from Llama 2 with longer training sequences and on a dataset where long texts are upsampled. We perform extensive evaluation on language modeling, synthetic context probing tasks, and a wide range of research benchmarks. On research benchmarks, our models achieve consistent improvements on most regular tasks and significant improvements on long-context tasks over Llama 2. Notably, with a cost-effective instruction tuning procedure that does not require human-annotated long instruction data, the 70B variant can already surpass gpt-3.5-turbo-16k's overall performance on a suite of long-context tasks. Alongside these results, we provide an in-depth analysis on the individual components of our method. We delve into Llama's position encodings and discuss its limitation in modeling long dependencies. We also examine the impact of various design choices in the pretraining process, including the data mix and the training curriculum of sequence lengths -- our ablation experiments suggest that having abundant long texts in the pretrain dataset is not the key to achieving strong performance, and we empirically verify that long context continual pretraining is more efficient and similarly effective compared to pretraining from scratch with long sequences.

Llama 2: Open Foundation and Fine-Tuned Chat Models

In this work, we develop and release Llama 2, a collection of pretrained and fine-tuned large language models (LLMs) ranging in scale from 7 billion to 70 billion parameters. Our fine-tuned LLMs, called Llama 2-Chat, are optimized for dialogue use cases. Our models outperform open-source chat models on most benchmarks we tested, and based on our human evaluations for helpfulness and safety, may be a suitable substitute for closed-source models. We provide a detailed description of our approach to fine-tuning and safety improvements of Llama 2-Chat in order to enable the community to build on our work and contribute to the responsible development of LLMs.

A Theory on Adam Instability in Large-Scale Machine Learning

We present a theory for the previously unexplained divergent behavior noticed in the training of large language models. We argue that the phenomenon is an artifact of the dominant optimization algorithm used for training, called Adam. We observe that Adam can enter a state in which the parameter update vector has a relatively large norm and is essentially uncorrelated with the direction of descent on the training loss landscape, leading to divergence. This artifact is more likely to be observed in the training of a deep model with a large batch size, which is the typical setting of large-scale language model training. To argue the theory, we present observations from the training runs of the language models of different scales: 7 billion, 30 billion, 65 billion, and 546 billion parameters.

UniMax: Fairer and more Effective Language Sampling for Large-Scale Multilingual Pretraining

Pretrained multilingual large language models have typically used heuristic temperature-based sampling to balance between different languages. However previous work has not systematically evaluated the efficacy of different pretraining language distributions across model scales. In this paper, we propose a new sampling method, UniMax, that delivers more uniform coverage of head languages while mitigating overfitting on tail languages by explicitly capping the number of repeats over each language's corpus. We perform an extensive series of ablations testing a range of sampling strategies on a suite of multilingual benchmarks, while varying model scale. We find that UniMax outperforms standard temperature-based sampling, and the benefits persist as scale increases. As part of our contribution, we release: (i) an improved and refreshed mC4 multilingual corpus consisting of 29 trillion characters across 107 languages, and (ii) a suite of pretrained umT5 model checkpoints trained with UniMax sampling.

Character-Aware Models Improve Visual Text Rendering

Current image generation models struggle to reliably produce well-formed visual text. In this paper, we investigate a key contributing factor: popular text-to-image models lack character-level input features, making it much harder to predict a word's visual makeup as a series of glyphs. To quantify the extent of this effect, we conduct a series of controlled experiments comparing character-aware vs. character-blind text encoders. In the text-only domain, we find that character-aware models provide large gains on a novel spelling task (WikiSpell). Transferring these learnings onto the visual domain, we train a suite of image generation models, and show that character-aware variants outperform their character-blind counterparts across a range of novel text rendering tasks (our DrawText benchmark). Our models set a much higher state-of-the-art on visual spelling, with 30+ point accuracy gains over competitors on rare words, despite training on far fewer examples.

FCM: Forgetful Causal Masking Makes Causal Language Models Better Zero-Shot Learners

Large language models (LLM) trained using the next-token-prediction objective, such as GPT3 and PaLM, have revolutionized natural language processing in recent years by showing impressive zero-shot and few-shot capabilities across a wide range of tasks. In this work, we propose a simple technique that significantly boosts the performance of LLMs without adding computational cost. Our key observation is that, by performing the next token prediction task with randomly selected past tokens masked out, we can improve the quality of the learned representations for downstream language understanding tasks. We hypothesize that randomly masking past tokens prevents over-attending to recent tokens and encourages attention to tokens in the distant past. By randomly masking input tokens in the PaLM model, we show that we can significantly improve 1B and 8B PaLM's zero-shot performance on the SuperGLUE benchmark from 55.7 to 59.2 and from 61.6 to 64.0, respectively. Our largest 8B model matches the score of PaLM with an average score of 64, despite the fact that PaLM is trained on a much larger dataset (780B tokens) of high-quality conversation and webpage data, while ours is trained on the smaller C4 dataset (180B tokens). Experimental results show that our method also improves PaLM's zero and few-shot performance on a diverse suite of tasks, including commonsense reasoning, natural language inference and cloze completion. Moreover, we show that our technique also helps representation learning, significantly improving PaLM's finetuning results.