Many existing benchmarks of large (multimodal) language models (LLMs) focus on measuring LLMs' academic proficiency, often with also an interest in comparing model performance with human test takers. While these benchmarks have proven key to the development of LLMs, they suffer from several limitations, including questionable measurement quality (e.g., Do they measure what they are supposed to in a reliable way?), lack of quality assessment on the item level (e.g., Are some items more important or difficult than others?) and unclear human population reference (e.g., To whom can the model be compared?). In response to these challenges, we propose leveraging knowledge from psychometrics - a field dedicated to the measurement of latent variables like academic proficiency - into LLM benchmarking. We make three primary contributions. First, we introduce PATCH: a novel framework for Psychometrics-AssisTed benCHmarking of LLMs. PATCH addresses the aforementioned limitations, presenting a new direction for LLM benchmark research. Second, we implement PATCH by measuring GPT-4 and Gemini-Pro-Vision's proficiency in 8th grade mathematics against 56 human populations. We show that adopting a psychometrics-based approach yields evaluation outcomes that diverge from those based on existing benchmarking practices. Third, we release 4 datasets to support measuring and comparing LLM proficiency in grade school mathematics and science against human populations.

Although commonsense reasoning is greatly shaped by cultural and geographical factors, previous studies on language models have predominantly centered on English cultures, potentially resulting in an Anglocentric bias. In this paper, we introduce IndoCulture, aimed at understanding the influence of geographical factors on language model reasoning ability, with a specific emphasis on the diverse cultures found within eleven Indonesian provinces. In contrast to prior works that relied on templates (Yin et al., 2022) and online scrapping (Fung et al., 2024), we created IndoCulture by asking local people to manually develop the context and plausible options based on predefined topics. Evaluations of 23 language models reveal several insights: (1) even the best open-source model struggles with an accuracy of 53.2%, (2) models often provide more accurate predictions for specific provinces, such as Bali and West Java, and (3) the inclusion of location contexts enhances performance, especially in larger models like GPT-4, emphasizing the significance of geographical context in commonsense reasoning.

Recent advancements in Large Language Models (LLMs) have significantly increased their presence in human-facing Artificial Intelligence (AI) applications. However, LLMs could reproduce and even exacerbate stereotypical outputs from training data. This work introduces the Multi-Grain Stereotype (MGS) dataset, encompassing 51,867 instances across gender, race, profession, religion, and stereotypical text, collected by fusing multiple previously publicly available stereotype detection datasets. We explore different machine learning approaches aimed at establishing baselines for stereotype detection, and fine-tune several language models of various architectures and model sizes, presenting in this work a series of stereotypes classifier models for English text trained on MGS. To understand whether our stereotype detectors capture relevant features (aligning with human common sense) we utilise a variety of explanainable AI tools, including SHAP, LIME, and BertViz, and analyse a series of example cases discussing the results. Finally, we develop a series of stereotype elicitation prompts and evaluate the presence of stereotypes in text generation tasks with popular LLMs, using one of our best performing previously presented stereotypes detectors. Our experiments yielded several key findings: i) Training stereotype detectors in a multi-dimension setting yields better results than training multiple single-dimension classifiers.ii) The integrated MGS Dataset enhances both the in-dataset and cross-dataset generalisation ability of stereotype detectors compared to using the datasets separately.

Constructing personalized and anthropomorphic agents holds significant importance in the simulation of social networks. However, there are still two key problems in existing works: the agent possesses world knowledge that does not belong to its personas, and it cannot eliminate the interference of diverse persona information on current actions, which reduces the personalization and anthropomorphism of the agent. To solve the above problems, we construct the social media agent based on personalized knowledge and dynamic persona information. For personalized knowledge, we add external knowledge sources and match them with the persona information of agents, thereby giving the agent personalized world knowledge. For dynamic persona information, we use current action information to internally retrieve the persona information of the agent, thereby reducing the interference of diverse persona information on the current action. To make the agent suitable for social media, we design five basic modules for it: persona, planning, action, memory and reflection. To provide an interaction and verification environment for the agent, we build a social media simulation sandbox. In the experimental verification, automatic and human evaluations demonstrated the effectiveness of the agent we constructed.

Online continual learning suffers from an underfitted solution due to insufficient training for prompt model update (e.g., single-epoch training). To address the challenge, we propose an efficient online continual learning method using the neural collapse phenomenon. In particular, we induce neural collapse to form a simplex equiangular tight frame (ETF) structure in the representation space so that the continuously learned model with a single epoch can better fit to the streamed data by proposing preparatory data training and residual correction in the representation space. With an extensive set of empirical validations using CIFAR-10/100, TinyImageNet, ImageNet-200, and ImageNet-1K, we show that our proposed method outperforms state-of-the-art methods by a noticeable margin in various online continual learning scenarios such as disjoint and Gaussian scheduled continuous (i.e., boundary-free) data setups.

Event detection is one of the fundamental tasks in information extraction and knowledge graph. However, a realistic event detection system often needs to deal with new event classes constantly. These new classes usually have only a few labeled instances as it is time-consuming and labor-intensive to annotate a large number of unlabeled instances. Therefore, this paper proposes a new task, called class-incremental few-shot event detection. Nevertheless, this task faces two problems, i.e., old knowledge forgetting and new class overfitting. To solve these problems, this paper further presents a novel knowledge distillation and prompt learning based method, called Prompt-KD. Specifically, to handle the forgetting problem about old knowledge, Prompt-KD develops an attention based multi-teacher knowledge distillation framework, where the ancestor teacher model pre-trained on base classes is reused in all learning sessions, and the father teacher model derives the current student model via adaptation. On the other hand, in order to cope with the few-shot learning scenario and alleviate the corresponding new class overfitting problem, Prompt-KD is also equipped with a prompt learning mechanism. Extensive experiments on two benchmark datasets, i.e., FewEvent and MAVEN, demonstrate the superior performance of Prompt-KD.

Automated generation of feedback on programming assignments holds significant benefits for programming education, especially when it comes to advanced assignments. Automated Program Repair techniques, especially Large Language Model based approaches, have gained notable recognition for their potential to fix introductory assignments. However, the programs used for evaluation are relatively simple. It remains unclear how existing approaches perform in repairing programs from higher-level programming courses. To address these limitations, we curate a new advanced student assignment dataset named Defects4DS from a higher-level programming course. Subsequently, we identify the challenges related to fixing bugs in advanced assignments. Based on the analysis, we develop a framework called PaR that is powered by the LLM. PaR works in three phases: Peer Solution Selection, Multi-Source Prompt Generation, and Program Repair. Peer Solution Selection identifies the closely related peer programs based on lexical, semantic, and syntactic criteria. Then Multi-Source Prompt Generation adeptly combines multiple sources of information to create a comprehensive and informative prompt for the last Program Repair stage. The evaluation on Defects4DS and another well-investigated ITSP dataset reveals that PaR achieves a new state-of-the-art performance, demonstrating impressive improvements of 19.94% and 15.2% in repair rate compared to prior state-of-the-art LLM- and symbolic-based approaches, respectively

Current foundation models exhibit impressive capabilities when prompted either with text only or with both image and text inputs. But do their capabilities change depending on the input modality? In this work, we propose $\textbf{IsoBench}$, a benchmark dataset containing problems from four major areas: math, science, algorithms, and games. Each example is presented with multiple $\textbf{isomorphic representations}$ of inputs, such as visual, textual, and mathematical presentations. IsoBench provides fine-grained feedback to diagnose performance gaps caused by the form of the representation. Across various foundation models, we observe that on the same problem, models have a consistent preference towards textual representations. Most prominently, when evaluated on all IsoBench problems, Claude-3 Opus performs 28.7 points worse when provided with images instead of text; similarly, GPT-4 Turbo is 18.7 points worse and Gemini Pro is 14.9 points worse. Finally, we present two prompting techniques, $\textit{IsoCombination}$ and $\textit{IsoScratchPad}$, which improve model performance by considering combinations of, and translations between, different input representations.

We use different methods to examine and explore the data and find differences in what students write about courses depending on gender of the examiner. Data from English and Swedish courses are evaluated and compared, in order to capture more nuance in the gender bias that might be found. Here we present the results from the work so far, but this is an ongoing project and there is more work to do.

Textual domain is a crucial property within the Natural Language Processing (NLP) community due to its effects on downstream model performance. The concept itself is, however, loosely defined and, in practice, refers to any non-typological property, such as genre, topic, medium or style of a document. We investigate the core notion of domains via human proficiency in identifying related intrinsic textual properties, specifically the concepts of genre (communicative purpose) and topic (subject matter). We publish our annotations in *TGeGUM*: A collection of 9.1k sentences from the GUM dataset (Zeldes, 2017) with single sentence and larger context (i.e., prose) annotations for one of 11 genres (source type), and its topic/subtopic as per the Dewey Decimal library classification system (Dewey, 1979), consisting of 10/100 hierarchical topics of increased granularity. Each instance is annotated by three annotators, for a total of 32.7k annotations, allowing us to examine the level of human disagreement and the relative difficulty of each annotation task. With a Fleiss' kappa of at most 0.53 on the sentence level and 0.66 at the prose level, it is evident that despite the ubiquity of domains in NLP, there is little human consensus on how to define them. By training classifiers to perform the same task, we find that this uncertainty also extends to NLP models.