Jimmy Huang

Md Tahmid Rahman Laskar, Sawsan Alqahtani, M Saiful Bari et al.

Large Language Models (LLMs) have recently gained significant attention due to their remarkable capabilities in performing diverse tasks across various domains. However, a thorough evaluation of these models is crucial before deploying them in real-world applications to ensure they produce reliable performance. Despite the well-established importance of evaluating LLMs in the community, the complexity of the evaluation process has led to varied evaluation setups, causing inconsistencies in findings and interpretations. To address this, we systematically review the primary challenges and limitations causing these inconsistencies and unreliable evaluations in various steps of LLM evaluation. Based on our critical review, we present our perspectives and recommendations to ensure LLM evaluations are reproducible, reliable, and robust.

Abeer Badawi, Md Tahmid Rahman Laskar, Elahe Rahimi et al.

The escalating global mental health crisis, marked by persistent treatment gaps, availability, and a shortage of qualified therapists, positions Large Language Models (LLMs) as a promising avenue for scalable support. While LLMs offer potential for accessible emotional assistance, their reliability, therapeutic relevance, and alignment with human standards remain challenging to address. This paper introduces a human-grounded evaluation methodology designed to assess LLM generated responses in therapeutic dialogue. Our approach involved curating a dataset of 500 mental health conversations from datasets with real-world scenario questions and evaluating the responses generated by nine diverse LLMs, including closed source and open source models. More specifically, these responses were evaluated by two psychiatric trained experts, who independently rated each on a 5 point Likert scale across a comprehensive 6 attribute rubric. This rubric captures Cognitive Support and Affective Resonance, providing a multidimensional perspective on therapeutic quality. Our analysis reveals that LLMs provide strong cognitive reliability by producing safe, coherent, and clinically appropriate information, but they demonstrate unstable affective alignment. Although closed source models (e.g., GPT-4o) offer balanced therapeutic responses, open source models show greater variability and emotional flatness. We reveal a persistent cognitive-affective gap and highlight the need for failure aware, clinically grounded evaluation frameworks that prioritize relational sensitivity alongside informational accuracy in mental health oriented LLMs. We advocate for balanced evaluation protocols with human in the loop that center on therapeutic sensitivity and provide a framework to guide the responsible design and clinical oversight of mental health oriented conversational AI.

Israt Jahan, Md Tahmid Rahman Laskar, Chun Peng et al.

Recently, Large Language Models (LLM) have demonstrated impressive capability to solve a wide range of tasks. However, despite their success across various tasks, no prior work has investigated their capability in the biomedical domain yet. To this end, this paper aims to evaluate the performance of LLMs on benchmark biomedical tasks. For this purpose, we conduct a comprehensive evaluation of 4 popular LLMs in 6 diverse biomedical tasks across 26 datasets. To the best of our knowledge, this is the first work that conducts an extensive evaluation and comparison of various LLMs in the biomedical domain. Interestingly, we find based on our evaluation that in biomedical datasets that have smaller training sets, zero-shot LLMs even outperform the current state-of-the-art fine-tuned biomedical models. This suggests that pretraining on large text corpora makes LLMs quite specialized even in the biomedical domain. We also find that not a single LLM can outperform other LLMs in all tasks, with the performance of different LLMs may vary depending on the task. While their performance is still quite poor in comparison to the biomedical models that were fine-tuned on large training sets, our findings demonstrate that LLMs have the potential to be a valuable tool for various biomedical tasks that lack large annotated data.

Israt Jahan, Md Tahmid Rahman Laskar, Chun Peng et al.

ChatGPT is a large language model developed by OpenAI. Despite its impressive performance across various tasks, no prior work has investigated its capability in the biomedical domain yet. To this end, this paper aims to evaluate the performance of ChatGPT on various benchmark biomedical tasks, such as relation extraction, document classification, question answering, and summarization. To the best of our knowledge, this is the first work that conducts an extensive evaluation of ChatGPT in the biomedical domain. Interestingly, we find based on our evaluation that in biomedical datasets that have smaller training sets, zero-shot ChatGPT even outperforms the state-of-the-art fine-tuned generative transformer models, such as BioGPT and BioBART. This suggests that ChatGPT's pre-training on large text corpora makes it quite specialized even in the biomedical domain. Our findings demonstrate that ChatGPT has the potential to be a valuable tool for various tasks in the biomedical domain that lack large annotated data.

Md Tahmid Rahman Laskar, Mizanur Rahman, Israt Jahan et al.

Debatepedia is a publicly available dataset consisting of arguments and counter-arguments on controversial topics that has been widely used for the single-document query-focused abstractive summarization task in recent years. However, it has been recently found that this dataset is limited by noise and even most queries in this dataset do not have any relevance to the respective document. In this paper, we present a methodology for cleaning the Debatepedia dataset by leveraging the generative power of large language models to make it suitable for query-focused abstractive summarization. More specifically, we harness the language generation capabilities of ChatGPT to regenerate its queries. We evaluate the effectiveness of the proposed ChatGPT annotated version of the Debatepedia dataset using several benchmark summarization models and demonstrate that the newly annotated version of Debatepedia outperforms the original dataset in terms of both query relevance as well as summary generation quality. We will make this annotated and cleaned version of the dataset publicly available.

Junpeng Zhao, Lin Li, Ming Li et al.

Modern data-driven recommendation systems risk memorizing sensitive user behavioral patterns, raising privacy concerns. Existing recommendation unlearning methods, while capable of removing target data influence, suffer from inefficient unlearning speed and degraded performance, failing to meet real-time unlearning demands. Considering the ranking-oriented nature of recommendation systems, we present unranking, the process of reducing the ranking positions of target items while ensuring the formal guarantees of recommendation unlearning. To achieve efficient unranking, we propose Learning to Fast Unrank in Collaborative Filtering Recommendation (L2UnRank), which operates through three key stages: (a) identifying the influenced scope via interaction-based p-hop propagation, (b) computing structural and semantic influences for entities within this scope, and (c) performing efficient, ranking-aware parameter updates guided by influence information. Extensive experiments across multiple datasets and backbone models demonstrate L2UnRank's model-agnostic nature, achieving state-of-the-art unranking effectiveness and maintaining recommendation quality comparable to retraining, while also delivering a 50x speedup over existing methods. Codes are available at https://github.com/Juniper42/L2UnRank.

Md Tahmid Rahman Laskar, Israt Jahan, Elham Dolatabadi et al.

Large Language Models (LLMs) have demonstrated impressive performance in biomedical relation extraction, even in zero-shot scenarios. However, evaluating LLMs in this task remains challenging due to their ability to generate human-like text, often producing synonyms or abbreviations of gold-standard answers, making traditional automatic evaluation metrics unreliable. On the other hand, while human evaluation is more reliable, it is costly and time-consuming, making it impractical for real-world applications. This paper investigates the use of LLMs-as-the-Judge as an alternative evaluation method for biomedical relation extraction. We benchmark 8 LLMs as judges to evaluate the responses generated by 5 other LLMs across 3 biomedical relation extraction datasets. Unlike other text-generation tasks, we observe that LLM-based judges perform quite poorly (usually below 50% accuracy) in the biomedical relation extraction task. Our findings reveal that it happens mainly because relations extracted by LLMs do not adhere to any standard format. To address this, we propose structured output formatting for LLM-generated responses that helps LLM-Judges to improve their performance by about 15% (on average). We also introduce a domain adaptation technique to further enhance LLM-Judge performance by effectively transferring knowledge between datasets. We release both our human-annotated and LLM-annotated judgment data (36k samples in total) for public use here: https://github.com/tahmedge/llm_judge_biomedical_re.

Md Tahmid Rahman Laskar, Mohammed Saidul Islam, Mir Tafseer Nayeem et al.

Automatic evaluators such as reward models play a central role in the alignment and evaluation of large vision-language models (LVLMs). Despite their growing importance, these evaluators are almost exclusively assessed on English-centric benchmarks, leaving open the question of how well these evaluators generalize across languages. To answer this question, we introduce MM-JudgeBench, the first large-scale benchmark for multilingual and multimodal judge model evaluation, which includes over 60K pairwise preference instances spanning 25 typologically diverse languages. MM-JudgeBench integrates two complementary subsets: a general vision-language preference evaluation subset extending VL-RewardBench, and a chart-centric visual-text reasoning subset derived from OpenCQA, enabling systematic analysis of reward models (i.e., LVLM judges) across diverse settings. We additionally release a multilingual training set derived from MM-RewardBench, disjoint from our evaluation data, to support domain adaptation. By evaluating 22 LVLMs (15 open-source, 7 proprietary), we uncover substantial cross-lingual performance variance in our proposed benchmark. Our analysis further shows that model size and architecture are poor predictors of multilingual robustness, and that even state-of-the-art LVLM judges exhibit inconsistent behavior across languages. Together, these findings expose fundamental limitations of current reward modeling and underscore the necessity of multilingual, multimodal benchmarks for developing reliable automated evaluators.

Israt Jahan, Md Tahmid Rahman Laskar, Chun Peng et al.

This paper presents a comprehensive evaluation of cost-efficient Large Language Models (LLMs) for diverse biomedical tasks spanning both text and image modalities. We evaluated a range of closed-source and open-source LLMs on tasks such as biomedical text classification and generation, question answering, and multimodal image processing. Our experimental findings indicate that there is no single LLM that can consistently outperform others across all tasks. Instead, different LLMs excel in different tasks. While some closed-source LLMs demonstrate strong performance on specific tasks, their open-source counterparts achieve comparable results (sometimes even better), with additional benefits like faster inference and enhanced privacy. Our experimental results offer valuable insights for selecting models that are optimally suited for specific biomedical applications.

Md Tahmid Rahman Laskar, Mohammed Saidul Islam, Ridwan Mahbub et al.

Charts are ubiquitous as they help people understand and reason with data. Recently, various downstream tasks, such as chart question answering, chart2text, and fact-checking, have emerged. Large Vision-Language Models (LVLMs) show promise in tackling these tasks, but their evaluation is costly and time-consuming, limiting real-world deployment. While using LVLMs as judges to assess the chart comprehension capabilities of other LVLMs could streamline evaluation processes, challenges like proprietary datasets, restricted access to powerful models, and evaluation costs hinder their adoption in industrial settings. To this end, we present a comprehensive evaluation of 13 open-source LVLMs as judges for diverse chart comprehension and reasoning tasks. We design both pairwise and pointwise evaluation tasks covering criteria like factual correctness, informativeness, and relevancy. Additionally, we analyze LVLM judges based on format adherence, positional consistency, length bias, and instruction-following. We focus on cost-effective LVLMs (<10B parameters) suitable for both research and commercial use, following a standardized evaluation protocol and rubric to measure the LVLM judge's accuracy. Experimental results reveal notable variability: while some open LVLM judges achieve GPT-4-level evaluation performance (about 80% agreement with GPT-4 judgments), others struggle (below ~10% agreement). Our findings highlight that state-of-the-art open-source LVLMs can serve as cost-effective automatic evaluators for chart-related tasks, though biases such as positional preference and length bias persist.

Abeer Badawi, Elahe Rahimi, Md Tahmid Rahman Laskar et al.

Evaluating Large Language Models (LLMs) for mental health support is challenging due to the emotionally and cognitively complex nature of therapeutic dialogue. Existing benchmarks are limited in scale, reliability, often relying on synthetic or social media data, and lack frameworks to assess when automated judges can be trusted. To address the need for large-scale dialogue datasets and judge reliability assessment, we introduce two benchmarks that provide a framework for generation and evaluation. MentalBench-100k consolidates 10,000 one-turn conversations from three real scenarios datasets, each paired with nine LLM-generated responses, yielding 100,000 response pairs. MentalAlign-70k}reframes evaluation by comparing four high-performing LLM judges with human experts across 70,000 ratings on seven attributes, grouped into Cognitive Support Score (CSS) and Affective Resonance Score (ARS). We then employ the Affective Cognitive Agreement Framework, a statistical methodology using intraclass correlation coefficients (ICC) with confidence intervals to quantify agreement, consistency, and bias between LLM judges and human experts. Our analysis reveals systematic inflation by LLM judges, strong reliability for cognitive attributes such as guidance and informativeness, reduced precision for empathy, and some unreliability in safety and relevance. Our contributions establish new methodological and empirical foundations for reliable, large-scale evaluation of LLMs in mental health. We release the benchmarks and codes at: https://github.com/abeerbadawi/MentalBench/

Yizheng Huang, Jimmy Huang

Retrieval-Augmented Generation (RAG) merges retrieval methods with deep learning advancements to address the static limitations of large language models (LLMs) by enabling the dynamic integration of up-to-date external information. This methodology, focusing primarily on the text domain, provides a cost-effective solution to the generation of plausible but possibly incorrect responses by LLMs, thereby enhancing the accuracy and reliability of their outputs through the use of real-world data. As RAG grows in complexity and incorporates multiple concepts that can influence its performance, this paper organizes the RAG paradigm into four categories: pre-retrieval, retrieval, post-retrieval, and generation, offering a detailed perspective from the retrieval viewpoint. It outlines RAG's evolution and discusses the field's progression through the analysis of significant studies. Additionally, the paper introduces evaluation methods for RAG, addressing the challenges faced and proposing future research directions. By offering an organized framework and categorization, the study aims to consolidate existing research on RAG, clarify its technological underpinnings, and highlight its potential to broaden the adaptability and applications of LLMs.

Yizheng Huang, Jimmy Huang

The rapid advancement of artificial intelligence (AI) has highlighted ChatGPT as a pivotal technology in the field of information retrieval (IR). Distinguished from its predecessors, ChatGPT offers significant benefits that have attracted the attention of both the industry and academic communities. While some view ChatGPT as a groundbreaking innovation, others attribute its success to the effective integration of product development and market strategies. The emergence of ChatGPT, alongside GPT-4, marks a new phase in Generative AI, generating content that is distinct from training examples and exceeding the capabilities of the prior GPT-3 model by OpenAI. Unlike the traditional supervised learning approach in IR tasks, ChatGPT challenges existing paradigms, bringing forth new challenges and opportunities regarding text quality assurance, model bias, and efficiency. This paper seeks to examine the impact of ChatGPT on IR tasks and offer insights into its potential future developments.

Md Tahmid Rahman Laskar, Mohammed Saidul Islam, Ridwan Mahbub et al.

Large Vision-Language Models (LVLMs) with only 7B parameters have shown promise as automated judges in chart comprehension tasks. However, tiny models (<=2B parameters) still perform poorly as judges, limiting their real-world use in resource-constrained settings. To address this, we propose two approaches to ensure cost-efficient evaluation: (i) multi-criteria prompting, which combines separate evaluation criteria into a single query, and (ii) domain-adaptive transfer learning, in which we fine-tune a 2B-parameter LVLM on synthetic judgments in a chart dataset to create the ChartJudge. Experiments show that multi-criteria prompting exposes robustness gaps, which led to a huge drop in performance for 7B models, including specialized LVLM judges like LLaVA-Critic. In addition, we find that our tiny LVLM (ChartJudge) can effectively transfer knowledge from one dataset to another to make it a more specialized model. Our fine-grained analysis across chart types and query complexities offers actionable insights into trade-offs between model size, prompt design, and transferability, enabling scalable, low-cost evaluation for chart reasoning tasks.

Md Tahmid Rahman Laskar, Jimmy Huang, Vladan Smetana et al.

Industrial Information Technology (IT) infrastructures are often vulnerable to cyberattacks. To ensure security to the computer systems in an industrial environment, it is required to build effective intrusion detection systems to monitor the cyber-physical systems (e.g., computer networks) in the industry for malicious activities. This paper aims to build such intrusion detection systems to protect the computer networks from cyberattacks. More specifically, we propose a novel unsupervised machine learning approach that combines the K-Means algorithm with the Isolation Forest for anomaly detection in industrial big data scenarios. Since our objective is to build the intrusion detection system for the big data scenario in the industrial domain, we utilize the Apache Spark framework to implement our proposed model which was trained in large network traffic data (about 123 million instances of network traffic) stored in Elasticsearch. Moreover, we evaluate our proposed model on the live streaming data and find that our proposed system can be used for real-time anomaly detection in the industrial setup. In addition, we address different challenges that we face while training our model on large datasets and explicitly describe how these issues were resolved. Based on our empirical evaluation in different use-cases for anomaly detection in real-world network traffic data, we observe that our proposed system is effective to detect anomalies in big data scenarios. Finally, we evaluate our proposed model on several academic datasets to compare with other models and find that it provides comparable performance with other state-of-the-art approaches.