Bridging biomolecular modeling with natural language information, particularly through large language models (LLMs), has recently emerged as a promising interdisciplinary research area. LLMs, having been trained on large corpora of scientific documents, demonstrate significant potential in understanding and reasoning about biomolecules by providing enriched contextual and domain knowledge. However, the extent to which LLM-driven insights can improve performance on complex predictive tasks (e.g., toxicity) remains unclear. Further, the extent to which relevant knowledge can be extracted from LLMs also remains unknown. In this study, we present Molecule Caption Arena: the first comprehensive benchmark of LLM-augmented molecular property prediction. We evaluate over twenty LLMs, including both general-purpose and domain-specific molecule captioners, across diverse prediction tasks. To this goal, we introduce a novel, battle-based rating system. Our findings confirm the ability of LLM-extracted knowledge to enhance state-of-the-art molecular representations, with notable model-, prompt-, and dataset-specific variations. Code, resources, and data are available at github.com/Genentech/molcap-arena.

Financial AI empowers sophisticated approaches to financial market forecasting, portfolio optimization, and automated trading. This survey provides a systematic analysis of these developments across three primary dimensions: predictive models that capture complex market dynamics, decision-making frameworks that optimize trading and investment strategies, and knowledge augmentation systems that leverage unstructured financial information. We examine significant innovations including foundation models for financial time series, graph-based architectures for market relationship modeling, and hierarchical frameworks for portfolio optimization. Analysis reveals crucial trade-offs between model sophistication and practical constraints, particularly in high-frequency trading applications. We identify critical gaps and open challenges between theoretical advances and industrial implementation, outlining open challenges and opportunities for improving both model performance and practical applicability.

Code-mixing (CM), where speakers blend languages within a single expression, is prevalent in multilingual societies but poses challenges for natural language processing due to its complexity and limited data. We propose using a large language model to generate synthetic CM data, which is then used to enhance the performance of task-specific models for CM sentiment analysis. Our results show that in Spanish-English, synthetic data improved the F1 score by 9.32%, outperforming previous augmentation techniques. However, in Malayalam-English, synthetic data only helped when the baseline was low; with strong natural data, additional synthetic data offered little benefit. Human evaluation confirmed that this approach is a simple, cost-effective way to generate natural-sounding CM sentences, particularly beneficial for low baselines. Our findings suggest that few-shot prompting of large language models is a promising method for CM data augmentation and has significant impact on improving sentiment analysis, an important element in the development of social influence systems.

Vision-language models, like CLIP (Contrastive Language Image Pretraining), are becoming increasingly popular for a wide range of multimodal retrieval tasks. However, prior work has shown that large language and deep vision models can learn historical biases contained in their training sets, leading to perpetuation of stereotypes and potential downstream harm. In this work, we conduct a systematic analysis of the social biases that are present in CLIP, with a focus on the interaction between image and text modalities. We first propose a taxonomy of social biases called So-B-IT, which contains 374 words categorized across ten types of bias. Each type can lead to societal harm if associated with a particular demographic group. Using this taxonomy, we examine images retrieved by CLIP from a facial image dataset using each word as part of a prompt. We find that CLIP frequently displays undesirable associations between harmful words and specific demographic groups, such as retrieving mostly pictures of Middle Eastern men when asked to retrieve images of a "terrorist". Finally, we conduct an analysis of the source of such biases, by showing that the same harmful stereotypes are also present in a large image-text dataset used to train CLIP models for examples of biases that we find. Our findings highlight the importance of evaluating and addressing bias in vision-language models, and suggest the need for transparency and fairness-aware curation of large pre-training datasets.

The conventional targeted adversarial attacks add a small perturbation to an image to make neural network models estimate the image as a predefined target class, even if it is not the correct target class. Recently, for visual-language models (VLMs), the focus of targeted adversarial attacks is to generate a perturbation that makes VLMs answer intended target text outputs. For example, they aim to make a small perturbation on an image to make VLMs' answers change from "there is an apple" to "there is a baseball." However, answering just intended text outputs is insufficient for tricky questions like "if there is a baseball, tell me what is below it." This is because the target of the adversarial attacks does not consider the overall integrity of the original image, thereby leading to a lack of visual reasoning. In this work, we focus on generating targeted adversarial examples with visual reasoning against VLMs. To this end, we propose 1) a novel adversarial attack procedure -- namely, Replace-then-Perturb and 2) a contrastive learning-based adversarial loss -- namely, Contrastive-Adv. In Replace-then-Perturb, we first leverage a text-guided segmentation model to find the target object in the image. Then, we get rid of the target object and inpaint the empty space with the desired prompt. By doing this, we can generate a target image corresponding to the desired prompt, while maintaining the overall integrity of the original image. Furthermore, in Contrastive-Adv, we design a novel loss function to obtain better adversarial examples. Our extensive benchmark results demonstrate that Replace-then-Perturb and Contrastive-Adv outperform the baseline adversarial attack algorithms. We note that the source code to reproduce the results will be available.

Pleasure and pain play an important role in human decision making by providing a common currency for resolving motivational conflicts. While Large Language Models (LLMs) can generate detailed descriptions of pleasure and pain experiences, it is an open question whether LLMs can recreate the motivational force of pleasure and pain in choice scenarios - a question which may bear on debates about LLM sentience, understood as the capacity for valenced experiential states. We probed this question using a simple game in which the stated goal is to maximise points, but where either the points-maximising option is said to incur a pain penalty or a non-points-maximising option is said to incur a pleasure reward, providing incentives to deviate from points-maximising behaviour. Varying the intensity of the pain penalties and pleasure rewards, we found that Claude 3.5 Sonnet, Command R+, GPT-4o, and GPT-4o mini each demonstrated at least one trade-off in which the majority of responses switched from points-maximisation to pain-minimisation or pleasure-maximisation after a critical threshold of stipulated pain or pleasure intensity is reached. LLaMa 3.1-405b demonstrated some graded sensitivity to stipulated pleasure rewards and pain penalties. Gemini 1.5 Pro and PaLM 2 prioritised pain-avoidance over points-maximisation regardless of intensity, while tending to prioritise points over pleasure regardless of intensity. We discuss the implications of these findings for debates about the possibility of LLM sentience.

Fact verification on tabular evidence incentivises the use of symbolic reasoning models where a logical form is constructed (e.g. a LISP-style program), providing greater verifiability than fully neural approaches. However, these systems typically rely on well-formed tables, restricting their use in many scenarios. An emerging symbolic reasoning paradigm for textual evidence focuses on natural logic inference, which constructs proofs by modelling set-theoretic relations between a claim and its evidence in natural language. This approach provides flexibility and transparency but is less compatible with tabular evidence since the relations do not extend to arithmetic functions. We propose a set-theoretic interpretation of numerals and arithmetic functions in the context of natural logic, enabling the integration of arithmetic expressions in deterministic proofs. We leverage large language models to generate arithmetic expressions by generating questions about salient parts of a claim which are answered by executing appropriate functions on tables. In a few-shot setting on FEVEROUS, we achieve an accuracy of 71.4, outperforming both fully neural and symbolic reasoning models by 3.4 points. When evaluated on TabFact without any further training, our method remains competitive with an accuracy lead of 0.5 points.

We present a light-weight approach for detecting nonfactual outputs from retrieval-augmented generation (RAG). Given a context and putative output, we compute a factuality score that can be thresholded to yield a binary decision to check the results of LLM-based question-answering, summarization, or other systems. Unlike factuality checkers that themselves rely on LLMs, we use compact, open-source natural language inference (NLI) models that yield a freely accessible solution with low latency and low cost at run-time, and no need for LLM fine-tuning. The approach also enables downstream mitigation and correction of hallucinations, by tracing them back to specific context chunks. Our experiments show high area under the ROC curve (AUC) across a wide range of relevant open source datasets, indicating the effectiveness of our method for fact-checking RAG output.

As human-agent teaming (HAT) research continues to grow, computational methods for modeling HAT behaviors and measuring HAT effectiveness also continue to develop. One rising method involves the use of human digital twins (HDT) to approximate human behaviors and socio-emotional-cognitive reactions to AI-driven agent team members. In this paper, we address three research questions relating to the use of digital twins for modeling trust in HATs. First, to address the question of how we can appropriately model and operationalize HAT trust through HDT HAT experiments, we conducted causal analytics of team communication data to understand the impact of empathy, socio-cognitive, and emotional constructs on trust formation. Additionally, we reflect on the current state of the HAT trust science to discuss characteristics of HAT trust that must be replicable by a HDT such as individual differences in trust tendencies, emergent trust patterns, and appropriate measurement of these characteristics over time. Second, to address the question of how valid measures of HDT trust are for approximating human trust in HATs, we discuss the properties of HDT trust: self-report measures, interaction-based measures, and compliance type behavioral measures. Additionally, we share results of preliminary simulations comparing different LLM models for generating HDT communications and analyze their ability to replicate human-like trust dynamics. Third, to address how HAT experimental manipulations will extend to human digital twin studies, we share experimental design focusing on propensity to trust for HDTs vs. transparency and competency-based trust for AI agents.

Retrieval-augmented generation (RAG) on specialized domain datasets has shown improved performance when large language models (LLMs) are fine-tuned for generating responses to user queries. In this study, we develop a cybersecurity question-answering (Q\&A) dataset, called AttackQA, and employ it to build a RAG-based Q\&A system designed for analysts in security operations centers. The dataset comprises 25,335 Q\&A pairs, accompanied by rationales to facilitate fine-tuning and evaluation. 80\% of the dataset was generated with help of a lightweight open-source LLM (LLama 3 8B), which produced over 1100 tokens per second with full 16-bit precision on SambaNova System's SN40L specialized hardware. To ensure dataset quality, we fine-tuned LLama 3 70B to detect and reject low-quality Q\&A pairs. In using the dataset for RAG, we demonstrate that fine-tuning open-source embeddings and LLMs can yield superior accuracy compared to OpenAI's state-of-the-art proprietary embedding and LLM (GPT-4o). Furthermore, we use Llama 3.1 405B as a judge to evaluate answer correctness, enabling the creation of a fully open-source, high-speed RAG and evaluation pipeline with a benchmark for model accuracy.