Bayesian Optimization is critically vulnerable to extreme outliers. Existing provably robust methods typically assume a bounded cumulative corruption budget, which makes them defenseless against even a single corruption of sufficient magnitude. To address this, we introduce a new adversary whose budget is only bounded in the frequency of corruptions, not in their magnitude. We then derive RCGP-UCB, an algorithm coupling the famous upper confidence bound (UCB) approach with a Robust Conjugate Gaussian Process (RCGP). We present stable and adaptive versions of RCGP-UCB, and prove that they achieve sublinear regret in the presence of up to $O(T^{1/2})$ and $O(T^{1/3})$ corruptions with possibly infinite magnitude. This robustness comes at near zero cost: without outliers, RCGP-UCB's regret bounds match those of the standard GP-UCB algorithm.

New Jersey's one million Asian American residents could determine the gubernatorial outcome as Democrats increase community outreach efforts.

Frontier AI systems, including large language models (LLMs), hold increasing influence over the epistemology of human users. Such influence can reinforce prevailing societal values, potentially contributing to the lock-in of misguided moral beliefs and, consequently, the perpetuation of problematic moral practices on a broad scale. We introduce progress alignment as a technical solution to mitigate this imminent risk. Progress alignment algorithms learn to emulate the mechanics of human moral progress, thereby addressing the susceptibility of existing alignment methods to contemporary moral blindspots.

Large Language Models (LLMs) currently struggle to sequentially add new memories and integrate new knowledge. These limitations contrast with the human ability to continuously learn from new experiences and acquire knowledge throughout life. Most existing approaches add memories either through large context windows or external memory buffers (e.g., Retrieval-Augmented Generation), and studies on knowledge injection rarely test scenarios resembling everyday life events. In this work, we introduce a continual learning framework, Memory Embedded in Gated LLMs (MEGa), which injects event memories directly into the weights of LLMs. Each memory is stored in a dedicated set of gated low-rank weights. During inference, a gating mechanism activates relevant memory weights by matching query embeddings to stored memory embeddings. This enables the model to both recall entire memories and answer related questions. On two datasets - fictional characters and Wikipedia events - MEGa outperforms baseline approaches in mitigating catastrophic forgetting. Our model draws inspiration from the complementary memory system of the human brain.

Artificial intelligence (AI) is transforming scientific research, including proteomics. Advances in mass spectrometry (MS)-based proteomics data quality, diversity, and scale, combined with groundbreaking AI techniques, are unlocking new challenges and opportunities in biological discovery. Here, we highlight key areas where AI is driving innovation, from data analysis to new biological insights. These include developing an AI-friendly ecosystem for proteomics data generation, sharing, and analysis; improving peptide and protein identification and quantification; characterizing protein-protein interactions and protein complexes; advancing spatial and perturbation proteomics; integrating multi-omics data; and ultimately enabling AI-empowered virtual cells.

Protein-ligand binding [Clyde et al., 2023] refers to the process as shown in Figure 1 by which ligands--usually small molecules, ions, or proteins--generate signals by binding to the active sites of target proteins through intermolecular forces. This binding typically changes the conformation of target proteins, which then results in the realization, modulation, or alteration of protein functions. Therefore, protein-ligand binding plays a central role in most, if not all, important life processes. For example, oxygen molecules are bound and carried through the human body by proteins like hemoglobin, and then utilized for energy production, while nonsteroidal anti-inflammatory drugs (NSAIDs) like ibuprofen work by inhibiting the functionality of the cyclooxygenase (COX) enzyme that thus reducing the release of pain-causing substances in the body. The concept and importance of binding affinity prediction were first addressed in Böhm [1994]: given the 3D structures of a target protein and a potential ligand, the objective is to predict the binding constant of such a complex, along with the most probable binding pose candidates. The prediction of the binding site (the set of protein residues that have at least one non-hydrogen atom within 4.0 Å of a ligand's non-hydrogen atom [Khazanov and Carlson, 2013]) and affinity (binding constants such as inhibition or dissociation constants, or the concentration at 50% inhibition) are usually divided into two separate but related stages [Ballester and Mitchell, 2010a]. One notable motivation for constructing a good binding affinity predictor (or scoring function, as called in some earlier work) is the essential role that it plays in drug discovery [Liu et al., 2023, 2024a] and virtual screening [Meng et al., 2011, Pinzi and Rastelli, 2019, Sadybekov and Katritch, 2023]. Traditional drug discovery essentially involves a process of trial and error.

Frontier AI systems, including large language models (LLMs), hold increasing influence over the epistemology of human users. Such influence can reinforce prevailing societal values, potentially contributing to the lock-in of misguided moral beliefs and, consequently, the perpetuation of problematic moral practices on a broad scale. We introduce progress alignment as a technical solution to mitigate this imminent risk. Progress alignment algorithms learn to emulate the mechanics of human moral progress, thereby addressing the susceptibility of existing alignment methods to contemporary moral blindspots. To empower research in progress alignment, we introduce ProgressGym, an experimental framework allowing the learning of moral progress mechanics from history, in order to facilitate future progress in real-world moral decisions. Leveraging 9 centuries of historical text and 18 historical LLMs, ProgressGym enables codification of real-world progress alignment challenges into concrete benchmarks. Specifically, we introduce three core challenges: tracking evolving values (PG-Follow), preemptively anticipating moral progress (PG-Predict), and regulating the feedback loop between human and AI value shifts (PG-Coevolve). Alignment methods without a temporal dimension are inapplicable to these tasks. In response, we present lifelong and extrapolative algorithms as baseline methods of progress alignment, and build an open leaderboard soliciting novel algorithms and challenges. The framework and the leaderboard are available at https://github.com/PKU-Alignment/ProgressGym and https://huggingface.co/spaces/PKU-Alignment/ProgressGym-LeaderBoard respectively.

Computing in the life sciences has undergone a transformative evolution, from early computational models in the 1950s to the applications of arti cial intelligence (AI) and machine learning (ML) seen today. This paper highlights key milestones and technological advancements through the historical development of computing in the life sciences. The discussion includes the inception of computational models for biological processes, the advent of bioinformatics tools, and the integration of AI/ML in modern life sciences research. Attention is given to AI-enabled tools used in the life sciences, such as scienti c large language models and bio-AI tools, examining their capabilities, limitations, and impact to biological risk. This paper seeks to clarify and establish essential terminology and concepts to ensure informed decision-making and e ective communication across disciplines. The views and opinions expressed within this manuscript are those of the authors and do not necessarily re ect the views and opinions of any organization the authors are a liated with.

The process of identifying a compound from its mass spectrum is a critical step in the analysis of complex mixtures. Typical solutions for the mass spectrum to compound (MS2C) problem involve matching the unknown spectrum against a library of known spectrum-molecule pairs, an approach that is limited by incomplete library coverage. Compound to mass spectrum (C2MS) models can improve retrieval rates by augmenting real libraries with predicted spectra. Unfortunately, many existing C2MS models suffer from problems with prediction resolution, scalability, or interpretability. We develop a new probabilistic method for C2MS prediction, FraGNNet, that can efficiently and accurately predict high-resolution spectra. FraGNNet uses a structured latent space to provide insight into the underlying processes that define the spectrum. Our model achieves state-of-the-art performance in terms of prediction error, and surpasses existing C2MS models as a tool for retrieval-based MS2C.