Researchers from MBZUAI have introduced SPECS, a new reference-free evaluation metric for long image captions that modifies CLIP to emphasize specificity. SPECS aims to improve the correlation with human judgment while maintaining computational efficiency compared to LLM-based metrics. The proposed approach is intended for iterative use during image captioning model development, offering a practical alternative to existing methods.
A new methodology emulating fact-checker criteria assesses news outlet factuality and bias using LLMs. The approach uses prompts based on fact-checking criteria to elicit and aggregate LLM responses for predictions. Experiments demonstrate improvements over baselines, with error analysis on media popularity and region, and a released dataset/code at https://github.com/mbzuai-nlp/llm-media-profiling.
This paper presents six experiments evaluating personalization and user tracking in web search engine results. The experiments involve comparing search results based on VPN location (including UAE vs others), logged-in status, network type, search engine, browser, and trained Google accounts. The study measures total hits, first hit, and correlation between hits to identify patterns of personalization. Why it matters: The findings shed light on the extent of filter bubble effects and potential biases in search results for users in the UAE and globally.
LAraBench introduces a benchmark for Arabic NLP and speech processing, evaluating LLMs like GPT-3.5-turbo, GPT-4, BLOOMZ, Jais-13b-chat, Whisper, and USM. The benchmark covers 33 tasks across 61 datasets, using zero-shot and few-shot learning techniques. Results show that SOTA models generally outperform LLMs in zero-shot settings, though larger LLMs with few-shot learning reduce the gap. Why it matters: This benchmark helps assess and improve the performance of LLMs on Arabic language tasks, highlighting areas where specialized models still excel.
This article discusses domain shift in machine learning, where testing data differs from training data, and methods to mitigate it via domain adaptation and generalization. Domain adaptation uses labeled source data and unlabeled target data. Domain generalization uses labeled data from single or multiple source domains to generalize to unseen target domains. Why it matters: Research in mitigating domain shift enhances the robustness and applicability of AI models in diverse real-world scenarios.