DHP Benchmark: Are LLMs Good NLG Evaluators?

Large Language Models (LLMs) are increasingly serving as evaluators in Natural Language Generation (NLG) tasks. However, the capabilities of LLMs in scoring NLG quality remain inadequately explored. Current studies depend on human assessments and simple metrics that fail to capture the discernment of LLMs across diverse NLG tasks. To address this gap, we propose the Discernment of Hierarchical Perturbation (DHP) benchmarking framework, which provides quantitative discernment scores for LLMs utilizing hierarchically perturbed text data and statistical tests to measure the NLG evaluation capabilities of LLMs systematically. We have re-established six evaluation datasets for this benchmark, covering four NLG tasks: Summarization, Story Completion, Question Answering, and Translation. Our comprehensive benchmarking of five major LLM series provides critical insight into their strengths and limitations as NLG evaluators.

CLR-Fact: Evaluating the Complex Logical Reasoning Capability of Large Language Models over Factual Knowledge

While large language models (LLMs) have demonstrated impressive capabilities across various natural language processing tasks by acquiring rich factual knowledge from their broad training data, their ability to synthesize and logically reason with this knowledge in complex ways remains underexplored. In this work, we present a systematic evaluation of state-of-the-art LLMs' complex logical reasoning abilities through a novel benchmark of automatically generated complex reasoning questions over general domain and biomedical knowledge graphs. Our extensive experiments, employing diverse in-context learning techniques, reveal that LLMs excel at reasoning over general world knowledge but face significant challenges with specialized domain-specific knowledge. We find that prompting with explicit Chain-of-Thought demonstrations can substantially improve LLM performance on complex logical reasoning tasks with diverse logical operations. Interestingly, our controlled evaluations uncover an asymmetry where LLMs display proficiency at set union operations, but struggle considerably with set intersections - a key building block of logical reasoning. To foster further work, we will publicly release our evaluation benchmark and code.

Exclusive Style Removal for Cross Domain Novel Class Discovery

As a promising field in open-world learning, \textit{Novel Class Discovery} (NCD) is usually a task to cluster unseen novel classes in an unlabeled set based on the prior knowledge of labeled data within the same domain. However, the performance of existing NCD methods could be severely compromised when novel classes are sampled from a different distribution with the labeled ones. In this paper, we explore and establish the solvability of NCD in cross domain setting with the necessary condition that style information must be removed. Based on the theoretical analysis, we introduce an exclusive style removal module for extracting style information that is distinctive from the baseline features, thereby facilitating inference. Moreover, this module is easy to integrate with other NCD methods, acting as a plug-in to improve performance on novel classes with different distributions compared to the seen labeled set. Additionally, recognizing the non-negligible influence of different backbones and pre-training strategies on the performance of the NCD methods, we build a fair benchmark for future NCD research. Extensive experiments on three common datasets demonstrate the effectiveness of our proposed module.

FairLENS: Assessing Fairness in Law Enforcement Speech Recognition

Automatic speech recognition (ASR) techniques have become powerful tools, enhancing efficiency in law enforcement scenarios. To ensure fairness for demographic groups in different acoustic environments, ASR engines must be tested across a variety of speakers in realistic settings. However, describing the fairness discrepancies between models with confidence remains a challenge. Meanwhile, most public ASR datasets are insufficient to perform a satisfying fairness evaluation. To address the limitations, we built FairLENS - a systematic fairness evaluation framework. We propose a novel and adaptable evaluation method to examine the fairness disparity between different models. We also collected a fairness evaluation dataset covering multiple scenarios and demographic dimensions. Leveraging this framework, we conducted fairness assessments on 1 open-source and 11 commercially available state-of-the-art ASR models. Our results reveal that certain models exhibit more biases than others, serving as a fairness guideline for users to make informed choices when selecting ASR models for a given real-world scenario. We further explored model biases towards specific demographic groups and observed that shifts in the acoustic domain can lead to the emergence of new biases.

Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context

In this report, we present the latest model of the Gemini family, Gemini 1.5 Pro, a highly compute-efficient multimodal mixture-of-experts model capable of recalling and reasoning over fine-grained information from millions of tokens of context, including multiple long documents and hours of video and audio. Gemini 1.5 Pro achieves near-perfect recall on long-context retrieval tasks across modalities, improves the state-of-the-art in long-document QA, long-video QA and long-context ASR, and matches or surpasses Gemini 1.0 Ultra's state-of-the-art performance across a broad set of benchmarks. Studying the limits of Gemini 1.5 Pro's long-context ability, we find continued improvement in next-token prediction and near-perfect retrieval (>99%) up to at least 10M tokens, a generational leap over existing models such as Claude 2.1 (200k) and GPT-4 Turbo (128k). Finally, we highlight surprising new capabilities of large language models at the frontier; when given a grammar manual for Kalamang, a language with fewer than 200 speakers worldwide, the model learns to translate English to Kalamang at a similar level to a person who learned from the same content.

Stabilizing Sharpness-aware Minimization Through A Simple Renormalization Strategy

Recently, sharpness-aware minimization (SAM) has attracted a lot of attention because of its surprising effectiveness in improving generalization performance.However, training neural networks with SAM can be highly unstable since the loss does not decrease along the direction of the exact gradient at the current point, but instead follows the direction of a surrogate gradient evaluated at another point nearby. To address this issue, we propose a simple renormalization strategy, dubbed StableSAM, so that the norm of the surrogate gradient maintains the same as that of the exact gradient. Our strategy is easy to implement and flexible enough to integrate with SAM and its variants, almost at no computational cost. With elementary tools from convex optimization and learning theory, we also conduct a theoretical analysis of sharpness-aware training, revealing that compared to stochastic gradient descent (SGD), the effectiveness of SAM is only assured in a limited regime of learning rate. In contrast, we show how StableSAM extends this regime of learning rate and when it can consistently perform better than SAM with minor modification. Finally, we demonstrate the improved performance of StableSAM on several representative data sets and tasks.

Abductive Logical Reasoning on Knowledge Graphs

Abductive reasoning is logical reasoning that makes educated guesses to infer the most likely reasons to explain the observations. However, the abductive logical reasoning over knowledge graphs (KGs) is underexplored in KG literature. In this paper, we initially and formally raise the task of abductive logical reasoning over KGs, which involves inferring the most probable logic hypothesis from the KGs to explain an observed entity set. Traditional approaches use symbolic methods, like searching, to tackle the knowledge graph problem. However, the symbolic methods are unsuitable for this task, because the KGs are naturally incomplete, and the logical hypotheses can be complex with multiple variables and relations. To address these issues, we propose a generative approach to create logical expressions based on observations. First, we sample hypothesis-observation pairs from the KG and use supervised training to train a generative model that generates hypotheses from observations. Since supervised learning only minimizes structural differences between generated and reference hypotheses, higher structural similarity does not guarantee a better explanation for observations. To tackle this issue, we introduce the Reinforcement Learning from the Knowledge Graph (RLF-KG) method, which minimizes the differences between observations and conclusions drawn from the generated hypotheses according to the KG. Experimental results demonstrate that transformer-based generative models can generate logical explanations robustly and efficiently. Moreover, with the assistance of RLF-KG, the generated hypothesis can provide better explanations for the observations, and the method of supervised learning with RLF-KG achieves state-of-the-art results on abductive knowledge graph reasoning on three widely used KGs.

Unsupervised Candidate Answer Extraction through Differentiable Masker-Reconstructor Model

Question generation is a widely used data augmentation approach with extensive applications, and extracting qualified candidate answers from context passages is a critical step for most question generation systems. However, existing methods for candidate answer extraction are reliant on linguistic rules or annotated data that face the partial annotation issue and challenges in generalization. To overcome these limitations, we propose a novel unsupervised candidate answer extraction approach that leverages the inherent structure of context passages through a Differentiable Masker-Reconstructor (DMR) Model with the enforcement of self-consistency for picking up salient information tokens. We curated two datasets with exhaustively-annotated answers and benchmark a comprehensive set of supervised and unsupervised candidate answer extraction methods. We demonstrate the effectiveness of the DMR model by showing its performance is superior among unsupervised methods and comparable to supervised methods.

Beyond Fairness: Age-Harmless Parkinson's Detection via Voice

Parkinson's disease (PD), a neurodegenerative disorder, often manifests as speech and voice dysfunction. While utilizing voice data for PD detection has great potential in clinical applications, the widely used deep learning models currently have fairness issues regarding different ages of onset. These deep models perform well for the elderly group (age $>$ 55) but are less accurate for the young group (age $\leq$ 55). Through our investigation, the discrepancy between the elderly and the young arises due to 1) an imbalanced dataset and 2) the milder symptoms often seen in early-onset patients. However, traditional debiasing methods are impractical as they typically impair the prediction accuracy for the majority group while minimizing the discrepancy. To address this issue, we present a new debiasing method using GradCAM-based feature masking combined with ensemble models, ensuring that neither fairness nor accuracy is compromised. Specifically, the GradCAM-based feature masking selectively obscures age-related features in the input voice data while preserving essential information for PD detection. The ensemble models further improve the prediction accuracy for the minority (young group). Our approach effectively improves detection accuracy for early-onset patients without sacrificing performance for the elderly group. Additionally, we propose a two-step detection strategy for the young group, offering a practical risk assessment for potential early-onset PD patients.

Efficient Single-Image Depth Estimation on Mobile Devices, Mobile AI & AIM 2022 Challenge: Report

Various depth estimation models are now widely used on many mobile and IoT devices for image segmentation, bokeh effect rendering, object tracking and many other mobile tasks. Thus, it is very crucial to have efficient and accurate depth estimation models that can run fast on low-power mobile chipsets. In this Mobile AI challenge, the target was to develop deep learning-based single image depth estimation solutions that can show a real-time performance on IoT platforms and smartphones. For this, the participants used a large-scale RGB-to-depth dataset that was collected with the ZED stereo camera capable to generated depth maps for objects located at up to 50 meters. The runtime of all models was evaluated on the Raspberry Pi 4 platform, where the developed solutions were able to generate VGA resolution depth maps at up to 27 FPS while achieving high fidelity results. All models developed in the challenge are also compatible with any Android or Linux-based mobile devices, their detailed description is provided in this paper.