Beyond correlation: The impact of human uncertainty in measuring the effectiveness of automatic evaluation and LLM-as-a-judge

The effectiveness of automatic evaluation of generative models is typically measured by comparing it to human evaluation using correlation metrics. However, metrics like Krippendorff's $\alpha$ and Randolph's $\kappa$, originally designed to measure the reliability of human labeling, make assumptions about human behavior and the labeling process. In this paper, we show how *relying on a single aggregate correlation score* can obscure fundamental differences between human behavior and automatic evaluation methods, including LLM-as-a-Judge. Specifically, we demonstrate that when the proportion of samples with variation or uncertainty in human labels (gathered during human evaluation) is relatively high, machine labels (generated by automatic evaluation methods) may superficially appear to have similar or better correlation with the human majority label compared to human-to-human (HH) correlation. This can create the misleading impression that automatic evaluation is accurate enough to approximate the human majority label. However, as the proportion of samples with consistent human labels increases, the correlation between machine labels and human majority labels declines, falling below HH correlation. Based on these findings, we first propose stratifying results by human label uncertainty to provide a more robust analysis of automatic evaluation performance. Second, recognizing that uncertainty and variation are inherent in perception-based human evaluations, such as those involving attitudes or preferences, we introduce a new metric - *binned Jensen-Shannon Divergence for perception* for such scenarios to better measure the effectiveness of automatic evaluations. Third, we present visualization techniques -- *perception charts*, to compare the strengths and limitations of automatic evaluation and to contextualize correlation measures appropriately

Salient Information Prompting to Steer Content in Prompt-based Abstractive Summarization

Large language models (LLMs) can generate fluent summaries across domains using prompting techniques, reducing the need to train models for summarization applications. However, crafting effective prompts that guide LLMs to generate summaries with the appropriate level of detail and writing style remains a challenge. In this paper, we explore the use of salient information extracted from the source document to enhance summarization prompts. We show that adding keyphrases in prompts can improve ROUGE F1 and recall, making the generated summaries more similar to the reference and more complete. The number of keyphrases can control the precision-recall trade-off. Furthermore, our analysis reveals that incorporating phrase-level salient information is superior to word- or sentence-level. However, the impact on hallucination is not universally positive across LLMs. To conduct this analysis, we introduce Keyphrase Signal Extractor (SigExt), a lightweight model that can be finetuned to extract salient keyphrases. By using SigExt, we achieve consistent ROUGE improvements across datasets and open-weight and proprietary LLMs without any LLM customization. Our findings provide insights into leveraging salient information in building prompt-based summarization systems.

CriSPO: Multi-Aspect Critique-Suggestion-guided Automatic Prompt Optimization for Text Generation

Large language models (LLMs) can generate fluent summaries across domains using prompting techniques, reducing the need to train models for summarization applications. However, crafting effective prompts that guide LLMs to generate summaries with the appropriate level of detail and writing style remains a challenge. In this paper, we explore the use of salient information extracted from the source document to enhance summarization prompts. We show that adding keyphrases in prompts can improve ROUGE F1 and recall, making the generated summaries more similar to the reference and more complete. The number of keyphrases can control the precision-recall trade-off. Furthermore, our analysis reveals that incorporating phrase-level salient information is superior to word- or sentence-level. However, the impact on hallucination is not universally positive across LLMs. To conduct this analysis, we introduce Keyphrase Signal Extractor (CriSPO), a lightweight model that can be finetuned to extract salient keyphrases. By using CriSPO, we achieve consistent ROUGE improvements across datasets and open-weight and proprietary LLMs without any LLM customization. Our findings provide insights into leveraging salient information in building prompt-based summarization systems.

MedDet: Generative Adversarial Distillation for Efficient Cervical Disc Herniation Detection

Cervical disc herniation (CDH) is a prevalent musculoskeletal disorder that significantly impacts health and requires labor-intensive analysis from experts. Despite advancements in automated detection of medical imaging, two significant challenges hinder the real-world application of these methods. First, the computational complexity and resource demands present a significant gap for real-time application. Second, noise in MRI reduces the effectiveness of existing methods by distorting feature extraction. To address these challenges, we propose three key contributions: Firstly, we introduced MedDet, which leverages the multi-teacher single-student knowledge distillation for model compression and efficiency, meanwhile integrating generative adversarial training to enhance performance. Additionally, we customize the second-order nmODE to improve the model's resistance to noise in MRI. Lastly, we conducted comprehensive experiments on the CDH-1848 dataset, achieving up to a 5% improvement in mAP compared to previous methods. Our approach also delivers over 5 times faster inference speed, with approximately 67.8% reduction in parameters and 36.9% reduction in FLOPs compared to the teacher model. These advancements significantly enhance the performance and efficiency of automated CDH detection, demonstrating promising potential for future application in clinical practice. See project website https://steve-zeyu-zhang.github.io/MedDet

Towards Generalizable Reinforcement Learning via Causality-Guided Self-Adaptive Representations

General intelligence requires quick adaption across tasks. While existing reinforcement learning (RL) methods have made progress in generalization, they typically assume only distribution changes between source and target domains. In this paper, we explore a wider range of scenarios where both the distribution and environment spaces may change. For example, in Atari games, we train agents to generalize to tasks with different levels of mode and difficulty, where there could be new state or action variables that never occurred in previous environments. To address this challenging setting, we introduce a causality-guided self-adaptive representation-based approach, called CSR, that equips the agent to generalize effectively and efficiently across a sequence of tasks with evolving dynamics. Specifically, we employ causal representation learning to characterize the latent causal variables and world models within the RL system. Such compact causal representations uncover the structural relationships among variables, enabling the agent to autonomously determine whether changes in the environment stem from distribution shifts or variations in space, and to precisely locate these changes. We then devise a three-step strategy to fine-tune the model under different scenarios accordingly. Empirical experiments show that CSR efficiently adapts to the target domains with only a few samples and outperforms state-of-the-art baselines on a wide range of scenarios, including our simulated environments, Cartpole, and Atari games.

Boosting Efficiency in Task-Agnostic Exploration through Causal Knowledge

The effectiveness of model training heavily relies on the quality of available training resources. However, budget constraints often impose limitations on data collection efforts. To tackle this challenge, we introduce causal exploration in this paper, a strategy that leverages the underlying causal knowledge for both data collection and model training. We, in particular, focus on enhancing the sample efficiency and reliability of the world model learning within the domain of task-agnostic reinforcement learning. During the exploration phase, the agent actively selects actions expected to yield causal insights most beneficial for world model training. Concurrently, the causal knowledge is acquired and incrementally refined with the ongoing collection of data. We demonstrate that causal exploration aids in learning accurate world models using fewer data and provide theoretical guarantees for its convergence. Empirical experiments, on both synthetic data and real-world applications, further validate the benefits of causal exploration.

Harnessing the Power of LLMs: Automating Unit Test Generation for High-Performance Computing

Unit testing is crucial in software engineering for ensuring quality. However, it's not widely used in parallel and high-performance computing software, particularly scientific applications, due to their smaller, diverse user base and complex logic. These factors make unit testing challenging and expensive, as it requires specialized knowledge and existing automated tools are often ineffective. To address this, we propose an automated method for generating unit tests for such software, considering their unique features like complex logic and parallel processing. Recently, large language models (LLMs) have shown promise in coding and testing. We explored the capabilities of Davinci (text-davinci-002) and ChatGPT (gpt-3.5-turbo) in creating unit tests for C++ parallel programs. Our results show that LLMs can generate mostly correct and comprehensive unit tests, although they have some limitations, such as repetitive assertions and blank test cases.

When Swarm Learning meets energy series data: A decentralized collaborative learning design based on blockchain

Machine learning models offer the capability to forecast future energy production or consumption and infer essential unknown variables from existing data. However, legal and policy constraints within specific energy sectors render the data sensitive, presenting technical hurdles in utilizing data from diverse sources. Therefore, we propose adopting a Swarm Learning (SL) scheme, which replaces the centralized server with a blockchain-based distributed network to address the security and privacy issues inherent in Federated Learning (FL)'s centralized architecture. Within this distributed Collaborative Learning framework, each participating organization governs nodes for inter-organizational communication. Devices from various organizations utilize smart contracts for parameter uploading and retrieval. Consensus mechanism ensures distributed consistency throughout the learning process, guarantees the transparent trustworthiness and immutability of parameters on-chain. The efficacy of the proposed framework is substantiated across three real-world energy series modeling scenarios with superior performance compared to Local Learning approaches, simultaneously emphasizing enhanced data security and privacy over Centralized Learning and FL method. Notably, as the number of data volume and the count of local epochs increases within a threshold, there is an improvement in model performance accompanied by a reduction in the variance of performance errors. Consequently, this leads to an increased stability and reliability in the outcomes produced by the model.

ConSiDERS-The-Human Evaluation Framework: Rethinking Human Evaluation for Generative Large Language Models

In this position paper, we argue that human evaluation of generative large language models (LLMs) should be a multidisciplinary undertaking that draws upon insights from disciplines such as user experience research and human behavioral psychology to ensure that the experimental design and results are reliable. The conclusions from these evaluations, thus, must consider factors such as usability, aesthetics, and cognitive biases. We highlight how cognitive biases can conflate fluent information and truthfulness, and how cognitive uncertainty affects the reliability of rating scores such as Likert. Furthermore, the evaluation should differentiate the capabilities and weaknesses of increasingly powerful large language models -- which requires effective test sets. The scalability of human evaluation is also crucial to wider adoption. Hence, to design an effective human evaluation system in the age of generative NLP, we propose the ConSiDERS-The-Human evaluation framework consisting of 6 pillars --Consistency, Scoring Critera, Differentiating, User Experience, Responsible, and Scalability.

Teaching Machines to Code: Smart Contract Translation with LLMs

The advent of large language models (LLMs) has marked a significant milestone in the realm of artificial intelligence, with their capabilities often matching or surpassing human expertise in various domains. Among these achievements, their adeptness in translation tasks stands out, closely mimicking the intricate and preliminary processes undertaken by human translators to ensure the fidelity and quality of the translated content. Despite the advancements in utilizing LLMs for translating programming code across different languages, the domain of smart contract translation, particularly into languages not previously encountered by the LLM, remains largely unexplored. In our research, we present a pioneering approach, SolMover, which harnesses the synergy of two distinct LLMs within a unified framework. This framework is designed to grasp coding principles and apply this understanding to the translation of code into an unfamiliar language. Our study delves into the capacity of LLMs to mimic human learning processes, offering an in-depth evaluation of our methodology for converting smart contracts written in Solidity to Move, a language with limited resources. The framework employs one LLM to decipher coding conventions for the new language, creating a blueprint for the second LLM, which, lacking planning abilities, possesses coding expertise. The empirical evidence from our experiments suggests that SolMover substantially enhances performance compared to gpt-3.5-turbo-1106, and achieves superior results over competitors such as Palm2 and Mixtral-8x7B-Instruct. Additionally, our analysis highlights the efficacy of our bug mitigation strategy in elevating code quality across all models, even outside the SolMover framework.