Decoding Biases: Automated Methods and LLM Judges for Gender Bias Detection in Language Models

Large Language Models (LLMs) have excelled at language understanding and generating human-level text. However, even with supervised training and human alignment, these LLMs are susceptible to adversarial attacks where malicious users can prompt the model to generate undesirable text. LLMs also inherently encode potential biases that can cause various harmful effects during interactions. Bias evaluation metrics lack standards as well as consensus and existing methods often rely on human-generated templates and annotations which are expensive and labor intensive. In this work, we train models to automatically create adversarial prompts to elicit biased responses from target LLMs. We present LLM- based bias evaluation metrics and also analyze several existing automatic evaluation methods and metrics. We analyze the various nuances of model responses, identify the strengths and weaknesses of model families, and assess where evaluation methods fall short. We compare these metrics to human evaluation and validate that the LLM-as-a-Judge metric aligns with human judgement on bias in response generation.

Zero-shot Conversational Summarization Evaluations with small Large Language Models

Large Language Models (LLMs) exhibit powerful summarization abilities. However, their capabilities on conversational summarization remains under explored. In this work we evaluate LLMs (approx. 10 billion parameters) on conversational summarization and showcase their performance on various prompts. We show that the summaries generated by models depend on the instructions and the performance of LLMs vary with different instructions sometimes resulting steep drop in ROUGE scores if prompts are not selected carefully. We also evaluate the models with human evaluations and discuss the limitations of the models on conversational summarization

Sample Efficient Multimodal Semantic Augmentation for Incremental Summarization

In this work, we develop a prompting approach for incremental summarization of task videos. We develop a sample-efficient few-shot approach for extracting semantic concepts as an intermediate step. We leverage an existing model for extracting the concepts from the images and extend it to videos and introduce a clustering and querying approach for sample efficiency, motivated by the recent advances in perceiver-based architectures. Our work provides further evidence that an approach with richer input context with relevant entities and actions from the videos and using these as prompts could enhance the summaries generated by the model. We show the results on a relevant dataset and discuss possible directions for the work.

Position Matters! Empirical Study of Order Effect in Knowledge-grounded Dialogue

With the power of large pretrained language models, various research works have integrated knowledge into dialogue systems. The traditional techniques treat knowledge as part of the input sequence for the dialogue system, prepending a set of knowledge statements in front of dialogue history. However, such a mechanism forces knowledge sets to be concatenated in an ordered manner, making models implicitly pay imbalanced attention to the sets during training. In this paper, we first investigate how the order of the knowledge set can influence autoregressive dialogue systems' responses. We conduct experiments on two commonly used dialogue datasets with two types of transformer-based models and find that models view the input knowledge unequally. To this end, we propose a simple and novel technique to alleviate the order effect by modifying the position embeddings of knowledge input in these models. With the proposed position embedding method, the experimental results show that each knowledge statement is uniformly considered to generate responses.

Human in the loop approaches in multi-modal conversational task guidance system development

Development of task guidance systems for aiding humans in a situated task remains a challenging problem. The role of search (information retrieval) and conversational systems for task guidance has immense potential to help the task performers achieve various goals. However, there are several technical challenges that need to be addressed to deliver such conversational systems, where common supervised approaches fail to deliver the expected results in terms of overall performance, user experience and adaptation to realistic conditions. In this preliminary work we first highlight some of the challenges involved during the development of such systems. We then provide an overview of existing datasets available and highlight their limitations. We finally develop a model-in-the-loop wizard-of-oz based data collection tool and perform a pilot experiment.

Distill and Collect for Semi-Supervised Temporal Action Segmentation

Recent temporal action segmentation approaches need frame annotations during training to be effective. These annotations are very expensive and time-consuming to obtain. This limits their performances when only limited annotated data is available. In contrast, we can easily collect a large corpus of in-domain unannotated videos by scavenging through the internet. Thus, this paper proposes an approach for the temporal action segmentation task that can simultaneously leverage knowledge from annotated and unannotated video sequences. Our approach uses multi-stream distillation that repeatedly refines and finally combines their frame predictions. Our model also predicts the action order, which is later used as a temporal constraint while estimating frames labels to counter the lack of supervision for unannotated videos. In the end, our evaluation of the proposed approach on two different datasets demonstrates its capability to achieve comparable performance to the full supervision despite limited annotation.

Controllable Response Generation for Assistive Use-cases

Conversational agents have become an integral part of the general population for simple task enabling situations. However, these systems are yet to have any social impact on the diverse and minority population, for example, helping people with neurological disorders, for example ALS, and people with speech, language and social communication disorders. Language model technology can play a huge role to help these users carry out daily communication and social interactions. To enable this population, we build a dialog system that can be controlled by users using cues or keywords. We build models that can suggest relevant cues in the dialog response context which is used to control response generation and can speed up communication. We also introduce a keyword loss to lexically constrain the model output. We show both qualitatively and quantitatively that our models can effectively induce the keyword into the model response without degrading the quality of response. In the context of usage of such systems for people with degenerative disorders, we present human evaluation of our cue or keyword predictor and the controllable dialog system and show that our models perform significantly better than models without control. Our study shows that keyword control on end to end response generation models is powerful and can enable and empower users with degenerative disorders to carry out their day to day communication.

Estimating Subjective Crowd-Evaluations as an Additional Objective to Improve Natural Language Generation

Human ratings are one of the most prevalent methods to evaluate the performance of natural language processing algorithms. Similarly, it is common to measure the quality of sentences generated by a natural language generation model using human raters. In this paper, we argue for exploring the use of subjective evaluations within the process of training language generation models in a multi-task learning setting. As a case study, we use a crowd-authored dialogue corpus to fine-tune six different language generation models. Two of these models incorporate multi-task learning and use subjective ratings of lines as part of an explicit learning goal. A human evaluation of the generated dialogue lines reveals that utterances generated by the multi-tasking models were subjectively rated as the most typical, most moving the conversation forward, and least offensive. Based on these promising first results, we discuss future research directions for incorporating subjective human evaluations into language model training and to hence keep the human user in the loop during the development process.

"Can you say more about the location?" The Development of a Pedagogical Reference Resolution Agent

In an increasingly globalized world, geographic literacy is crucial. In this paper, we present a collaborative two-player game to improve people's ability to locate countries on the world map. We discuss two implementations of the game: First, we created a web-based version which can be played with the remote-controlled agent Nellie. With the knowledge we gained from a large online data collection, we re-implemented the game so it can be played face-to-face with the Furhat robot Neil. Our analysis shows that participants found the game not just engaging to play, they also believe they gained lasting knowledge about the world map.

A System for Automated Image Editing from Natural Language Commands

This work presents the task of modifying images in an image editing program using natural language written commands. We utilize a corpus of over 6000 image edit text requests to alter real world images collected via crowdsourcing. A novel framework composed of actions and entities to map a user's natural language request to executable commands in an image editing program is described. We resolve previously labeled annotator disagreement through a voting process and complete annotation of the corpus. We experimented with different machine learning models and found that the LSTM, the SVM, and the bidirectional LSTM-CRF joint models are the best to detect image editing actions and associated entities in a given utterance.