Currently deployed public-key cryptosystems will be vulnerable to attacks by full-scale quantum computers. Consequently, quantum resistant cryptosystems are in high demand, and lattice-based cryptosystems, based on a hard problem known as Learning With Errors (LWE), have emerged as strong contenders for standardization. In this work, we train transformers to perform modular arithmetic and mix half-trained models and statistical cryptanalysis techniques to propose SALSA: a machine learning attack on LWE-based cryptographic schemes. SALSA can fully recover secrets for small-to-mid size LWE instances with sparse binary secrets, and may scale to attack real world LWE-based cryptosystems.

The looming threat of quantum computers has upended the field of cryptography.

Despite their impressive generalization capabilities, instruction-tuned Large Language Models often underperform on text classification benchmarks. We introduce SALSA, a coherent pipeline that combines structured prompting, class-to-token mapping, and parameter-efficient fine-tuning, thereby avoiding cold-start training. Each class label is mapped to a distinct output token, and prompts are constructed to elicit a single-token response. During inference, the model's output is projected only onto the logits of the relevant class tokens, enabling efficient and accurate classification in a single forward pass. SALSA achieves state-of-the-art results across diverse benchmarks, demonstrating its robustness and scalability for LLM-based classification applications.

The advent of quantum computing threatens classical public-key cryptography, motivating NIST's adoption of post-quantum schemes such as those based on the Module Learning With Errors (Module-LWE) problem. We present NoMod ML-Attack, a hybrid white-box cryptanalytic method that circumvents the challenge of modeling modular reduction by treating wrap-arounds as statistical corruption and casting secret recovery as robust linear estimation. Our approach combines optimized lattice preprocessing--including reduced-vector saving and algebraic amplification--with robust estimators trained via Tukey's Biweight loss. Experiments show NoMod achieves full recovery of binary secrets for dimension n = 350, recovery of sparse binomial secrets for n = 256, and successful recovery of sparse secrets in CRYST ALS-Kyber settings with parameters (n, k) = (128, 3) and (256, 2). We release our implementation in an anonymous repository https://anonymous.4open.science/r/NoMod-3BD4. The dawn of quantum computing presents a significant and growing threat to current cryptographic systems, many of which may be vulnerable to decryption through quantum-based attacks. At the heart of this risk is Shor's algorithm, a quantum-based algorithm developed in 1994 by Peter Shor, which can efficiently factor large integers and compute discrete logarithms. These two mathematical problems are computationally challenging for classical computers when the input size is large. In particular, while classical algorithms to factor integers, such as the General Number Field Sieve (GNFS), run in sub-exponential time, Shor's algorithm could run in polynomial time, when implemented on a sufficiently robust quantum computer Shor (1994; 1997). This development poses a significant threat to the security assumptions underlying widely used public-key cryptographic schemes, such as RSA, Elliptic Curve Cryptography (ECC), and the Diffie-Hellman key exchange. These algorithms are central to the Public Key Infrastructure (PKI) that secures virtually all modern digital communications.

A.1 Ring Learning with Errors ( 2) We now define RLWE samples and explain how to get LWE instances from them. We give a proof of the search binary-LWE to decisional binary-LWE reduction. Moreover, there are also attacks that do not use lattice reduction. Binary and ternary secret distributions are widely used in homomorphic encryption schemes. Let us now turn to the attacks on (sparse) binary/ternary secrets.

The looming threat of quantum computers has upended the field of cryptography.

Exhaustive virtual screening is highly informative but often intractable against the expensive objective functions involved in modern drug discovery. This problem is exacerbated in combinatorial contexts such as multi-vector expansion, where molecular spaces can quickly become ultra-large. Here, we introduce Scalable Active Learning via Synthon Acquisition (SALSA): a simple algorithm applicable to multi-vector expansion which extends pool-based active learning to non-enumerable spaces by factoring modeling and acquisition over synthon or fragment choices. Through experiments on ligand-and structure-based objectives, we highlight SALSA's sample efficiency, and its ability to scale to spaces of trillions of compounds. Further, we demonstrate application toward multi-parameter objective design tasks on three protein targets - finding SALSA-generated molecules have comparable chemical property profiles to known bioactives, and exhibit greater diversity and higher scores over an industry-leading generative approach. Given the strong association between a molecule's core scaffold and its chemical properties, a common workflow is to iteratively design, make, and test changes at targeted R-groups in order to advance therapeutics through the discovery pipeline (Schneider, 2017). Exhaustive virtual screening of R-group changes aids designers and medicinal chemists in the search for promising, synthesizable molecular structures, but quickly becomes intractable against computationally expensive scores as the number of possible attachments increases.

Currently deployed public-key cryptosystems will be vulnerable to attacks by full-scale quantum computers. Consequently, "quantum resistant" cryptosystems are in high demand, and lattice-based cryptosystems, based on a hard problem known as Learning With Errors (LWE), have emerged as strong contenders for standardization. In this work, we train transformers to perform modular arithmetic and mix half-trained models and statistical cryptanalysis techniques to propose SALSA: a machine learning attack on LWE-based cryptographic schemes. SALSA can fully recover secrets for small-to-mid size LWE instances with sparse binary secrets, and may scale to attack real world LWE-based cryptosystems.

This report is the system description of the MaLei team (Manchester and Leiden) for shared task Plain Language Adaptation of Biomedical Abstracts (PLABA) 2024 (we had an earlier name BeeManc following last year). This report contains two sections corresponding to the two sub-tasks in PLABA 2024. In task one, we applied fine-tuned ReBERTa-Base models to identify and classify the difficult terms, jargon and acronyms in the biomedical abstracts and reported the F1 score. Due to time constraints, we didn't finish the replacement task. In task two, we leveraged Llamma3.1-70B-Instruct and GPT-4o with the one-shot prompts to complete the abstract adaptation and reported the scores in BLEU, SARI, BERTScore, LENS, and SALSA. From the official Evaluation from PLABA-2024 on Task 1A and 1B, our \textbf{much smaller fine-tuned RoBERTa-Base} model ranked 3rd and 2nd respectively on the two sub-task, and the \textbf{1st on averaged F1 scores across the two tasks} from 9 evaluated systems. Our LLaMA-3.1-70B-instructed model achieved the \textbf{highest Completeness} score for Task-2. We share our fine-tuned models and related resources at \url{https://github.com/HECTA-UoM/PLABA2024}

In Large Language Model (LLM) development, Reinforcement Learning from Human Feedback (RLHF) is crucial for aligning models with human values and preferences. RLHF traditionally relies on the Kullback-Leibler (KL) divergence between the current policy and a frozen initial policy as a reference, which is added as a penalty in policy optimization algorithms like Proximal Policy Optimization (PPO). While this constraint prevents models from deviating too far from the initial checkpoint, it limits exploration of the reward landscape, reducing the model's ability to discover higher-quality solutions. As a result, policy optimization is often trapped in a narrow region of the parameter space, leading to suboptimal alignment and performance. This paper presents SALSA (Soup-based Alignment Learning for Stronger Adaptation), a novel approach designed to overcome these limitations by creating a more flexible and better located reference model through weight-space averaging of two independent supervised fine-tuned (SFT) models. This model soup allows for larger deviation in KL divergence and exploring a promising region of the solution space without sacrificing stability.