This work aims toward an application of natural language processing (NLP) technology for automatic verification of causal graphs using text sources. A causal graph is often derived from unsupervised causal discovery methods and requires manual evaluation from human experts. NLP technologies, i.e., Large Language Models (LLMs) such as BERT and ChatGPT, can potentially be used to verify the resulted causal graph by predicting if causal relation can be observed between node pairs based on the textual context. In this work, we compare the performance of two types of NLP models: (1) Pre-trained language models fine-tuned for causal relation classification task and, (2) prompt-based LLMs. Contrasted to previous studies where prompt-based LLMs work relatively well over a set of diverse tasks, preliminary experiments on biomedical and open-domain datasets suggest that the fine-tuned models far outperform the prompt-based LLMs, up to 20.5 points improvement of F1 score. We shared the code and the pre-processed datasets in our repository.

Evaluating startups in their early stages is a complex task that requires detailed analysis by experts. While automating this process on a large scale can significantly impact businesses, the inherent complexity poses challenges. This paper addresses this challenge by introducing the Startup Success Forecasting Framework (SSFF), a new automated system that combines traditional machine learning with advanced language models. This intelligent agent-based architecture is designed to reason, act, synthesize, and decide like a venture capitalist to perform the analysis end-to-end. The SSFF is made up of three main parts: - Prediction Block: Uses random forests and neural networks to make predictions. - Analyst Block: Simulates VC analysis scenario and uses SOTA prompting techniques - External Knowledge Block: Gathers real-time information from external sources. This framework requires minimal input data about the founder and startup description, enhances it with additional data from external resources, and performs a detailed analysis with high accuracy, all in an automated manner

Large Language Models (LLMs) are becoming vital tools that help us solve and understand complex problems by acting as digital assistants. LLMs can generate convincing explanations, even when only given the inputs and outputs of these problems, i.e., in a ``black-box'' approach. However, our research uncovers a hidden risk tied to this approach, which we call *adversarial helpfulness*. This happens when an LLM's explanations make a wrong answer look right, potentially leading people to trust incorrect solutions. In this paper, we show that this issue affects not just humans, but also LLM evaluators. Digging deeper, we identify and examine key persuasive strategies employed by LLMs. Our findings reveal that these models employ strategies such as reframing the questions, expressing an elevated level of confidence, and cherry-picking evidence to paint misleading answers in a credible light. To examine if LLMs are able to navigate complex-structured knowledge when generating adversarially helpful explanations, we create a special task based on navigating through graphs. Most LLMs are not able to find alternative paths along simple graphs, indicating that their misleading explanations aren't produced by only logical deductions using complex knowledge. These findings shed light on the limitations of the black-box explanation setting and allow us to provide advice on the safe usage of LLMs.

The increasing need for competent computing graduates proficient in programming, software development, and related technical competencies [Ca17] is one of the factors exacerbating pressure on higher education institutions to offer high quality, competency-based education [Ra21]. However, the latter requires extensive resources, mentoring, and, for example, formative feedback for learners, especially in introductory programming classes [Je22; Lo24]. This is due to the fact that novices experience a number of challenges in the process, which have been subject to extensive research in the past decades [Du86; Lu18; SS86]. Among them are cognitively demanding competencies [Ki20; Ki24], such as problem understanding, designing and writing algorithms, debugging, and understanding error messages [Du86; ER16; Ki20; Lu18; SS86]). Educators' expectations towards novice learners and what they can achieve in their first semester(s) seem to be too high and unrealistic [Lu16; Lu18; WCL07]. Moreover, the student-educator ratio in introductory programming classes keeps increasing in German higher education institutions, thereby limiting resources to provide feedback and hints, and adequately address heterogeneous prior knowledge and diverse educational biographies [Pe16; SB22].

Automated clinical text anonymization has the potential to unlock the widespread sharing of textual health data for secondary usage while assuring patient privacy and safety. Despite the proposal of many complex and theoretically successful anonymization solutions in literature, these techniques remain flawed. As such, clinical institutions are still reluctant to apply them for open access to their data. Recent advances in developing Large Language Models (LLMs) pose a promising opportunity to further the field, given their capability to perform various tasks. This paper proposes six new evaluation metrics tailored to the challenges of generative anonymization with LLMs. Moreover, we present a comparative study of LLM-based methods, testing them against two baseline techniques. Our results establish LLM-based models as a reliable alternative to common approaches, paving the way toward trustworthy anonymization of clinical text.

The current electroencephalogram (EEG) based deep learning models are typically designed for specific datasets and applications in brain-computer interaction (BCI), limiting the scale of the models and thus diminishing their perceptual capabilities and generalizability. Recently, Large Language Models (LLMs) have achieved unprecedented success in text processing, prompting us to explore the capabilities of Large EEG Models (LEMs). We hope that LEMs can break through the limitations of different task types of EEG datasets, and obtain universal perceptual capabilities of EEG signals through unsupervised pre-training. Then the models can be fine-tuned for different downstream tasks. However, compared to text data, the volume of EEG datasets is generally small and the format varies widely. For example, there can be mismatched numbers of electrodes, unequal length data samples, varied task designs, and low signal-to-noise ratio. To overcome these challenges, we propose a unified foundation model for EEG called Large Brain Model (LaBraM). LaBraM enables cross-dataset learning by segmenting the EEG signals into EEG channel patches. Vector-quantized neural spectrum prediction is used to train a semantically rich neural tokenizer that encodes continuous raw EEG channel patches into compact neural codes. We then pre-train neural Transformers by predicting the original neural codes for the masked EEG channel patches. The LaBraMs were pre-trained on about 2,500 hours of various types of EEG signals from around 20 datasets and validated on multiple different types of downstream tasks. Experiments on abnormal detection, event type classification, emotion recognition, and gait prediction show that our LaBraM outperforms all compared SOTA methods in their respective fields. Electroencephalography (EEG) is a method to record an electrogram of the spontaneous electrical activity of the brain.

Machine unlearning empowers individuals with the `right to be forgotten' by removing their private or sensitive information encoded in machine learning models. However, it remains uncertain whether MU can be effectively applied to Multimodal Large Language Models (MLLMs), particularly in scenarios of forgetting the leaked visual data of concepts. To overcome the challenge, we propose an efficient method, Single Image Unlearning (SIU), to unlearn the visual recognition of a concept by fine-tuning a single associated image for few steps. SIU consists of two key aspects: (i) Constructing Multifaceted fine-tuning data. We introduce four targets, based on which we construct fine-tuning data for the concepts to be forgotten; (ii) Jointly training loss. To synchronously forget the visual recognition of concepts and preserve the utility of MLLMs, we fine-tune MLLMs through a novel Dual Masked KL-divergence Loss combined with Cross Entropy loss. Alongside our method, we establish MMUBench, a new benchmark for MU in MLLMs and introduce a collection of metrics for its evaluation. Experimental results on MMUBench show that SIU completely surpasses the performance of existing methods. Furthermore, we surprisingly find that SIU can avoid invasive membership inference attacks and jailbreak attacks. To the best of our knowledge, we are the first to explore MU in MLLMs. We will release the code and benchmark in the near future.

Large language models (LLM) have emerged as a powerful tool for AI, with the key ability of in-context learning (ICL), where they can perform well on unseen tasks based on a brief series of task examples without necessitating any adjustments to the model parameters. One recent interesting mysterious observation is that models of different scales may have different ICL behaviors: larger models tend to be more sensitive to noise in the test context. This work studies this observation theoretically aiming to improve the understanding of LLM and ICL. We analyze two stylized settings: (1) linear regression with one-layer single-head linear transformers and (2) parity classification with two-layer multiple attention heads transformers (non-linear data and non-linear model). In both settings, we give closed-form optimal solutions and find that smaller models emphasize important hidden features while larger ones cover more hidden features; thus, smaller models are more robust to noise while larger ones are more easily distracted, leading to different ICL behaviors. This sheds light on where transformers pay attention to and how that affects ICL. Preliminary experimental results on large base and chat models provide positive support for our analysis.

Recent advancements in large multimodal language models have demonstrated remarkable proficiency across a wide range of tasks. Yet, these models still struggle with understanding the nuances of human humor through juxtaposition, particularly when it involves nonlinear narratives that underpin many jokes and humor cues. This paper investigates this challenge by focusing on comics with contradictory narratives, where each comic consists of two panels that create a humorous contradiction. We introduce the YesBut benchmark, which comprises tasks of varying difficulty aimed at assessing AI's capabilities in recognizing and interpreting these comics, ranging from literal content comprehension to deep narrative reasoning. Through extensive experimentation and analysis of recent commercial or open-sourced large (vision) language models, we assess their capability to comprehend the complex interplay of the narrative humor inherent in these comics. Our results show that even state-of-the-art models still lag behind human performance on this task. Our findings offer insights into the current limitations and potential improvements for AI in understanding human creative expressions.

Large language models are usually fine-tuned to align with human preferences. However, fine-tuning a large language model can be challenging. In this work, we introduce weak-to-strong search, framing the alignment of a large language model as a test-time greedy search to maximize the log-likelihood difference between small tuned and untuned models while sampling from the frozen large model. This method serves both as (i) a compute-efficient model up-scaling strategy that avoids directly tuning the large model and as (ii) an instance of weak-to-strong generalization that enhances a strong model with weak test-time guidance. Empirically, we demonstrate the flexibility of weak-to-strong search across different tasks. In controlled-sentiment generation and summarization, we use tuned and untuned gpt2s to effectively improve the alignment of large models without additional training. Crucially, in a more difficult instruction-following benchmark, AlpacaEval 2.0, we show that reusing off-the-shelf small models (e.g., zephyr-7b-beta and its untuned version) can significantly improve the length-controlled win rates of both white-box and black-box large models against gpt-4-turbo (e.g., 34.4 37.9 for Llama-3-70B-Instruct and 16.0 20.1 for gpt-3.5-turbo-instruct),