Text-to-image diffusion models often exhibit biases toward specific demographic groups, such as generating more males than females when prompted to generate images of engineers, raising ethical concerns and limiting their adoption. In this paper, we tackle the challenge of mitigating generation bias towards any target attribute value (e.g., “male” for “gender”) in diffusion models while preserving generation quality. We propose FairGen, an adaptive latent guidance mechanism which controls the generation distribution during inference. In FairGen, a latent guidance module dynamically adjusts the diffusion process to enforce specific attributes, while a memory module tracks the generation statistics and steers latent guidance to align with the targeted fair distribution of the attribute values. Further, given the limitations of existing datasets in comprehensively assessing bias in diffusion models, we introduce a holistic bias evaluation benchmark HBE, covering diverse domains and incorporating complex prompts across various applications. Extensive evaluations on HBE and Stable Bias datasets demonstrate that FairGen outperforms existing bias mitigation approaches, achieving substantial bias reduction (e.g., 68.5% gender bias reduction on Stable Diffusion 2). Ablation studies highlight FairGen’s ability to flexibly and precisely control generation distribution at any user-specified granularity, ensuring adaptive and targeted bias mitigation.

Large Language Models (LLMs) pruning seeks to remove unimportant weights for inference speedup with minimal accuracy impact. However, existing methods often suffer from accuracy degradation without full-model sparsity-aware fine-tuning. This paper presents Wanda++, a novel pruning framework that outperforms the state-of-the-art methods by utilizing decoder-block-level regional gradients. Specifically, Wanda++ improves the pruning score with regional gradients for the first time and proposes an efficient regional optimization method to minimize pruning-induced output discrepancies between the dense and sparse decoder output. Notably, Wanda++ improves perplexity by up to 32% over Wanda in the language modeling task and generalizes effectively to downstream tasks. Moreover, despite updating weights with regional optimization, Wanda++ remains orthogonal to sparsity-aware fine-tuning, further reducing perplexity with LoRA in great extend. Our approach is lightweight, pruning a 7B LLaMA model in under 10 minutes on a single H100 GPU.

As the use of Large Language Models (LLMs) becomes more widespread, understanding their self-evaluation of confidence in generated responses becomes increasingly important as it is integral to the reliability of the output of these models. We introduce the concept of Confidence-Probability Alignment, that connects an LLM’s internal confidence, quantified by token probabilities, to the confidence conveyed in the model’s response when explicitly asked about its certainty. Using various datasets and prompting techniques that encourage model introspection, we probe the alignment between models’ internal and expressed confidence. These techniques encompass using structured evaluation scales to rate confidence, including answer options when prompting, and eliciting the model’s confidence level for outputs it does not recognize as its own. Notably, among the models analyzed, OpenAI’s GPT-4 showed the strongest confidence-probability alignment, with an average Spearman’s  ̂𝜌 of 0.42, across a wide range of tasks. Our work contributes to the ongoing efforts to facilitate risk assessment in the application of LLMs and to further our understanding of model trustworthiness.

Research on Large Language Models (LLMs) has often neglected subtle biases that, although less apparent, can significantly influence the models’ outputs toward particular social narratives. This study addresses two such biases within LLMs: representative bias, which denotes a tendency of LLMs to generate outputs that mirror the experiences of certain identity groups, and affinity bias, reflecting the models’ evaluative preferences for specific narratives or viewpoints. We introduce two novel metrics to measure these biases: the Representative Bias Score (RBS) and the Affinity Bias Score (ABS), and present the Creativity-Oriented Generation Suite (CoGS), a collection of open-ended tasks such as short story writing and poetry composition, designed with customized rubrics to detect these subtle biases. Our analysis uncovers marked representative biases in prominent LLMs, with a preference for identities associated with being white, straight, and men. Furthermore, our investigation of affinity bias reveals distinctive evaluative patterns within each model, akin to ‘bias fingerprints’. This trend is also seen in human evaluators, highlighting a complex interplay between human and machine bias perceptions.

Large language models (LLMs), even when specifically trained to process long input contexts, struggle to capture relevant information located in the middle of their input. This phenomenon has been known as the lost-in-the-middle problem. In this work, we make three contributions. First, we set out to understand the factors that cause this phenomenon. In doing so, we establish a connection between lost-in-the-middle to LLMs’ intrinsic attention bias: LLMs exhibit an U-shaped attention bias where the tokens at the beginning and at the end of its input receive higher attention, regardless of their relevance. Second, we mitigate this positional bias through a calibration mechanism, found-in-the-middle, that allows the model to attend to contexts faithfully according to their relevance, even though when they are in the middle. Third, we show found-in-the-middle not only achieves better performance in locating relevant information within a long context, but also eventually leads to improved retrieval-augmented generation (RAG) performance across various tasks, outperforming existing methods by up to 10 percentage point. These findings open up future directions in understanding LLM attention bias and its potential consequences.

Social media networks have become a space where users are free to relate their opinions and sentiments which may lead to a large spreading of hatred or abusive messages which have to be moderated. This paper proposes a supervised approach to hate speech detection from a multilingual perspective. We focus in particular on hateful messages towards two different targets (immigrants and women) in English tweets, as well as sexist messages in both English and French. Several models have been developed ranging from feature-engineering approaches to neural ones. Our experiments show very encouraging results on both languages.

The massive growth of user-generated web content through blogs, online forums and most notably, social media networks, led to a large spreading of hatred or abusive messages which have to be moderated. This paper proposes a supervised approach to hate speech detection towards immigrants and women in English tweets. Several models have been developed ranging from feature-engineering approaches to neural ones.

In this paper, we propose a language-agnostic deep neural network architecture for aspect-based sentiment analysis. The proposed approach is based on Bidirectional Long Short-Term Memory (Bi-LSTM) network, which is further assisted with extra hand-crafted features. We define three different architectures for the successful combination of word embeddings and hand-crafted features. We evaluate the proposed approach for six languages (i.e. English, Spanish, French, Dutch, German and Hindi) and two problems (i.e. aspect term extraction and aspect sentiment classification). Experiments show that the proposed model attains state-of-the-art performance in most of the settings.

We propose a novel two-layered attention network based on Bidirectional Long Short-Term Memory for sentiment analysis. The novel two-layered attention network takes advantage of the external knowledge bases to improve the sentiment prediction. It uses the Knowledge Graph Embedding generated using the WordNet. We build our model by combining the two-layered attention network with the supervised model based on Support Vector Regression using a Multilayer Perceptron network for sentiment analysis. We evaluate our model on the benchmark dataset of SemEval 2017 Task 5. Experimental results show that the proposed model surpasses the top system of SemEval 2017 Task 5. The model performs significantly better by improving the state-of-the-art system at SemEval 2017 Task 5 by 1.7 and 3.7 points for sub-tracks 1 and 2 respectively.

In this paper, we propose a novel method for combining deep learning and classical feature based models using a Multi-Layer Perceptron (MLP) network for financial sentiment analysis. We develop various deep learning models based on Convolutional Neural Network (CNN), Long Short Term Memory (LSTM) and Gated Recurrent Unit (GRU). These are trained on top of pre-trained, autoencoder-based, financial word embeddings and lexicon features. An ensemble is constructed by combining these deep learning models and a classical supervised model based on Support Vector Regression (SVR). We evaluate our proposed technique on a benchmark dataset of SemEval-2017 shared task on financial sentiment analysis. The propose model shows impressive results on two datasets, i.e. microblogs and news headlines datasets. Comparisons show that our proposed model performs better than the existing state-of-the-art systems for the above two datasets by 2.0 and 4.1 cosine points, respectively.

This paper reports team IITPB’s participation in the SemEval 2017 Task 5 on ‘Fine-grained sentiment analysis on financial microblogs and news’. We developed 2 systems for the two tracks. One system was based on an ensemble of Support Vector Classifier and Logistic Regression. This system relied on Distributional Thesaurus (DT), word embeddings and lexicon features to predict a floating sentiment value between -1 and +1. The other system was based on Support Vector Regression using word embeddings, lexicon features, and PMI scores as features. The system was ranked 5th in track 1 and 8th in track 2.