In Artificial Intelligence, there is an increasing demand for adaptive models capable of dealing with a diverse spectrum of learning tasks, surpassing the limitations of systems devised to cope with a single task. The recent emergence of General-Purpose Artificial Intelligence Systems (GPAIS) poses model configuration and adaptability challenges at far greater complexity scales than the optimal design of traditional Machine Learning models. Evolutionary Computation (EC) has been a useful tool for both the design and optimization of Machine Learning models, endowing them with the capability to configure and/or adapt themselves to the task under consideration. Therefore, their application to GPAIS is a natural choice. This paper aims to analyze the role of EC in the field of GPAIS, exploring the use of EC for their design or enrichment. We also match GPAIS properties to Machine Learning areas in which EC has had a notable contribution, highlighting recent milestones of EC for GPAIS. Furthermore, we discuss the challenges of harnessing the benefits of EC for GPAIS, presenting different strategies to both design and improve GPAIS with EC, covering tangential areas, identifying research niches, and outlining potential research directions for EC and GPAIS.

Large language models manifest the ability of few-shot adaptation to a sequence of provided examples. This behavior, known as in-context learning, allows for performing nontrivial machine learning tasks during inference only. In this work, we address the question: can we leverage in-context learning to predict out-of-distribution materials properties? However, this would not be possible for structure property prediction tasks unless an effective method is found to pass atomic-level geometric features to the transformer model. To address this problem, we employ a compound model in which GPT-2 acts on the output of geometry-aware graph neural networks to adapt in-context information. To demonstrate our model's capabilities, we partition the QM9 dataset into sequences of molecules that share a common substructure and use them for in-context learning. This approach significantly improves the performance of the model on out-of-distribution examples, surpassing the one of general graph neural network models.

Large language models (LLMs) with billions of parameters demonstrate impressive performance. However, the widely used Multi-Head Attention (MHA) in LLMs incurs substantial computational and memory costs during inference. While some efforts have optimized attention mechanisms by pruning heads or sharing parameters among heads, these methods often lead to performance degradation or necessitate substantial continued pre-training costs to restore performance. Based on the analysis of attention redundancy, we design a Decoupled-Head Attention (DHA) mechanism. DHA adaptively configures group sharing for key heads and value heads across various layers, achieving a better balance between performance and efficiency. Inspired by the observation of clustering similar heads, we propose to progressively transform the MHA checkpoint into the DHA model through linear fusion of similar head parameters step by step, retaining the parametric knowledge of the MHA checkpoint. We construct DHA models by transforming various scales of MHA checkpoints given target head budgets. Our experiments show that DHA remarkably requires a mere 0.25\% of the original model's pre-training budgets to achieve 97.6\% of performance while saving 75\% of KV cache. Compared to Group-Query Attention (GQA), DHA achieves a 5$\times$ training acceleration, a maximum of 13.93\% performance improvement under 0.01\% pre-training budget, and 4\% relative improvement under 0.05\% pre-training budget.

Virtually all federated learning (FL) methods, including FedAvg, operate in the following manner: i) an orchestrating server sends the current model parameters to a cohort of clients selected via certain rule, ii) these clients then independently perform a local training procedure (e.g., via SGD or Adam) using their own training data, and iii) the resulting models are shipped to the server for aggregation. This process is repeated until a model of suitable quality is found. A notable feature of these methods is that each cohort is involved in a single communication round with the server only. In this work we challenge this algorithmic design primitive and investigate whether it is possible to ``squeeze more juice" out of each cohort than what is possible in a single communication round. Surprisingly, we find that this is indeed the case, and our approach leads to up to 74% reduction in the total communication cost needed to train a FL model in the cross-device setting. Our method is based on a novel variant of the stochastic proximal point method (SPPM-AS) which supports a large collection of client sampling procedures some of which lead to further gains when compared to classical client selection approaches.

As Large Language Models (LLMs) are increasingly deployed in real-world settings, understanding the knowledge they implicitly use when making decisions is critical. One way to capture this knowledge is in the form of Bayesian prior distributions. We develop a prompt-based workflow for eliciting prior distributions from LLMs. Our approach is based on iterated learning, a Markov chain Monte Carlo method in which successive inferences are chained in a way that supports sampling from the prior distribution. We validated our method in settings where iterated learning has previously been used to estimate the priors of human participants -- causal learning, proportion estimation, and predicting everyday quantities. We found that priors elicited from GPT-4 qualitatively align with human priors in these settings. We then used the same method to elicit priors from GPT-4 for a variety of speculative events, such as the timing of the development of superhuman AI.

Evaluating large language models (LLMs) is challenging. Traditional ground-truth-based benchmarks fail to capture the comprehensiveness and nuance of real-world queries, while LLM-as-judge benchmarks suffer from grading biases and limited query quantity. Both of them may also become contaminated over time. User-facing evaluation, such as Chatbot Arena, provides reliable signals but is costly and slow. In this work, we propose MixEval, a new paradigm for establishing efficient, gold-standard LLM evaluation by strategically mixing off-the-shelf benchmarks. It bridges (1) comprehensive and well-distributed real-world user queries and (2) efficient and fairly-graded ground-truth-based benchmarks, by matching queries mined from the web with similar queries from existing benchmarks. Based on MixEval, we further build MixEval-Hard, which offers more room for model improvement. Our benchmarks' advantages lie in (1) a 0.96 model ranking correlation with Chatbot Arena arising from the highly impartial query distribution and grading mechanism, (2) fast, cheap, and reproducible execution (6% of the time and cost of MMLU), and (3) dynamic evaluation enabled by the rapid and stable data update pipeline. We provide extensive meta-evaluation and analysis for our and existing LLM benchmarks to deepen the community's understanding of LLM evaluation and guide future research directions.

Feature learning has long been considered to be a major advantage of neural networks. However, how gradient-based training algorithms can learn useful features is not well-understood. In particular, the most widely applied analysis for overparametrized neural networks is the neural tangent kernel(NTK)(Jacot et al., 2018; Du et al., 2019; Allen-Zhu et al., 2019b). In this setting, the neurons don't move far from their initialization and the features are determined by the network architecture and random initialization (Chizat et al., 2019). While there are empirical and theoretical evidence on the limitation of NTK regime (Chizat et al., 2019; Arora et al., 2019), extending the analysis beyond the NTK regime has been challenging. For 2-layer networks, an alternative framework for analyzing overparametrized neural networks called mean-field analysis was introduced. Earlier mean-field analysis (e.g., Chizat and Bach, 2018; Mei et al., 2018) require either infinite or exponentially many neurons. Later works (e.g., Li et al., 2020; Ge et al., 2021; Bietti et al., 2022; Mahankali et al., 2024) can analyze the training dynamics of mildly overparametrized networks with polynomially many neurons with stronger assumptions on the ground-truth function.

The advent of large language models (LLMs) has dramatically advanced the state-of-the-art in numerous natural language generation tasks. For LLMs to be applied reliably, it is essential to have an accurate measure of their confidence. Currently, the most commonly used confidence score function is the likelihood of the generated sequence, which, however, conflates semantic and syntactic components. For instance, in question-answering (QA) tasks, an awkward phrasing of the correct answer might result in a lower probability prediction. Additionally, different tokens should be weighted differently depending on the context. In this work, we propose enhancing the predicted sequence probability by assigning different weights to various tokens using attention values elicited from the base LLM. By employing a validation set, we can identify the relevant attention heads, thereby significantly improving the reliability of the vanilla sequence probability confidence measure. We refer to this new score as the Contextualized Sequence Likelihood (CSL). CSL is easy to implement, fast to compute, and offers considerable potential for further improvement with task-specific prompts. Across several QA datasets and a diverse array of LLMs, CSL has demonstrated significantly higher reliability than state-of-the-art baselines in predicting generation quality, as measured by the AUROC or AUARC.

Retrieval augmented generation has revolutionized large language model (LLM) outputs by providing factual supports. Nevertheless, it struggles to capture all the necessary knowledge for complex reasoning questions. Existing retrieval methods typically divide reference documents into passages, treating them in isolation. These passages, however, are often interrelated, such as passages that are contiguous or share the same keywords. Therefore, recognizing the relatedness is crucial for enhancing the retrieval process. In this paper, we propose a novel retrieval method, called GNN-Ret, which leverages graph neural networks (GNNs) to enhance retrieval by considering the relatedness between passages. Specifically, we first construct a graph of passages by connecting passages that are structure-related and keyword-related. A graph neural network (GNN) is then leveraged to exploit the relationships between passages and improve the retrieval of supporting passages. Furthermore, we extend our method to handle multi-hop reasoning questions using a recurrent graph neural network (RGNN), named RGNN-Ret. At each step, RGNN-Ret integrates the graphs of passages from previous steps, thereby enhancing the retrieval of supporting passages. Extensive experiments on benchmark datasets demonstrate that GNN-Ret achieves higher accuracy for question answering with a single query of LLMs than strong baselines that require multiple queries, and RGNN-Ret further improves accuracy and achieves state-of-the-art performance, with up to 10.4% accuracy improvement on the 2WikiMQA dataset.

Epigraphs Stone inscriptions are important artifacts in the field of archaeology. Although several cultures follow different methods as primary forms of writing, for example, palm leaf manuscripts by Dravidians, papyrus manuscripts by Egyptians, and Animal Hide manuscripts by several European civilizations, stone inscriptions remained a robust secondary form of writing across all the civilizations that practiced writing. This is mainly due to the robustness associated with the medium, as it is impossible to manipulate, change the stone inscriptions and very difficult to mutilate them. Therefore, several historically important documents such as treaties, grants, and tombstones are engraved in stones.