Abstract
Many pretrained multilingual models exhibit cross-lingual transfer ability, which is often attributed to a learned language-neutral representation during pretraining. However, it remains unclear what factors contribute to the learning of a language-neutral representation, and whether the learned language-neutral representation suffices to facilitate cross-lingual transfer. We propose a synthetic task, Multilingual Othello (mOthello), as a testbed to delve into these two questions. We find that: (1) models trained with naive multilingual pretraining fail to learn a language-neutral representation across all input languages; (2) the introduction of "anchor tokens" (i.e., lexical items that are identical across languages) helps cross-lingual representation alignment; and (3) the learning of a language-neutral representation alone is not sufficient to facilitate cross-lingual transfer. Based on our findings, we propose a novel approach - multilingual pretraining with unified output space - that both induces the learning of language-neutral representation and facilitates cross-lingual transfer.
Beyond Human Norms: Unveiling Unique Values of Large Language Models through Interdisciplinary Approaches
Abstract
Recent advancements in Large Language Models (LLMs) have revolutionized the AI field but also pose potential safety and ethical risks. Deciphering LLMs' embedded values becomes crucial for assessing and mitigating their risks. Despite extensive investigation into LLMs' values, previous studies heavily rely on human-oriented value systems in social sciences. Then, a natural question arises: Do LLMs possess unique values beyond those of humans? Delving into it, this work proposes a novel framework, ValueLex, to reconstruct LLMs' unique value system from scratch, leveraging psychological methodologies from human personality/value research. Based on Lexical Hypothesis, ValueLex introduces a generative approach to elicit diverse values from 30+ LLMs, synthesizing a taxonomy that culminates in a comprehensive value framework via factor analysis and semantic clustering. We identify three core value dimensions, Competence, Character, and Integrity, each with specific subdimensions, revealing that LLMs possess a structured, albeit non-human, value system. Based on this system, we further develop tailored projective tests to evaluate and analyze the value inclinations of LLMs across different model sizes, training methods, and data sources. Our framework fosters an interdisciplinary paradigm of understanding LLMs, paving the way for future AI alignment and regulation.
Keyword: aligning
Rethinking the Evaluation of Dialogue Systems: Effects of User Feedback on Crowdworkers and LLMs
Authors: Clemencia Siro, Mohammad Aliannejadi, Maarten de Rijke
Subjects: Information Retrieval (cs.IR); Computation and Language (cs.CL)
Abstract
In ad-hoc retrieval, evaluation relies heavily on user actions, including implicit feedback. In a conversational setting such signals are usually unavailable due to the nature of the interactions, and, instead, the evaluation often relies on crowdsourced evaluation labels. The role of user feedback in annotators' assessment of turns in a conversational perception has been little studied. We focus on how the evaluation of task-oriented dialogue systems (TDSs), is affected by considering user feedback, explicit or implicit, as provided through the follow-up utterance of a turn being evaluated. We explore and compare two methodologies for assessing TDSs: one includes the user's follow-up utterance and one without. We use both crowdworkers and large language models (LLMs) as annotators to assess system responses across four aspects: relevance, usefulness, interestingness, and explanation quality. Our findings indicate that there is a distinct difference in ratings assigned by both annotator groups in the two setups, indicating user feedback does influence system evaluation. Workers are more susceptible to user feedback on usefulness and interestingness compared to LLMs on interestingness and relevance. User feedback leads to a more personalized assessment of usefulness by workers, aligning closely with the user's explicit feedback. Additionally, in cases of ambiguous or complex user requests, user feedback improves agreement among crowdworkers. These findings emphasize the significance of user feedback in refining system evaluations and suggest the potential for automated feedback integration in future research. We publicly release the annotated data to foster research in this area.
Keyword: align
BIRD: A Trustworthy Bayesian Inference Framework for Large Language Models
Abstract
Large language models primarily rely on inductive reasoning for decision making. This results in unreliable decisions when applied to real-world tasks that often present incomplete contexts and conditions. Thus, accurate probability estimation and appropriate interpretations are required to enhance decision-making reliability. In this paper, we propose a Bayesian inference framework called BIRD for large language models. BIRD provides controllable and interpretable probability estimation for model decisions, based on abductive factors, LLM entailment, as well as learnable deductive Bayesian modeling. Experiments show that BIRD produces probability estimations that align with human judgments over 65% of the time using open-sourced Llama models, outperforming the state-of-the-art GPT-4 by 35%. We also show that BIRD can be directly used for trustworthy decision making on many real-world applications.
Data Alignment for Zero-Shot Concept Generation in Dermatology AI
Authors: Soham Gadgil, Mahtab Bigverdi
Subjects: Computer Vision and Pattern Recognition (cs.CV); Computation and Language (cs.CL); Machine Learning (cs.LG)
Abstract
AI in dermatology is evolving at a rapid pace but the major limitation to training trustworthy classifiers is the scarcity of data with ground-truth concept level labels, which are meta-labels semantically meaningful to humans. Foundation models like CLIP providing zero-shot capabilities can help alleviate this challenge by leveraging vast amounts of image-caption pairs available on the internet. CLIP can be fine-tuned using domain specific image-caption pairs to improve classification performance. However, CLIP's pre-training data is not well-aligned with the medical jargon that clinicians use to perform diagnoses. The development of large language models (LLMs) in recent years has led to the possibility of leveraging the expressive nature of these models to generate rich text. Our goal is to use these models to generate caption text that aligns well with both the clinical lexicon and with the natural human language used in CLIP's pre-training data. Starting with captions used for images in PubMed articles, we extend them by passing the raw captions through an LLM fine-tuned on the field's several textbooks. We find that using captions generated by an expressive fine-tuned LLM like GPT-3.5 improves downstream zero-shot concept classification performance.
Abstract
Do vision-language models (VLMs) pre-trained to caption an image of a "durian" learn visual concepts such as "brown" (color) and "spiky" (texture) at the same time? We aim to answer this question as visual concepts learned "for free" would enable wide applications such as neuro-symbolic reasoning or human-interpretable object classification. We assume that the visual concepts, if captured by pre-trained VLMs, can be extracted by their vision-language interface with text-based concept prompts. We observe that recent works prompting VLMs with concepts often differ in their strategies to define and evaluate the visual concepts, leading to conflicting conclusions. We propose a new concept definition strategy based on two observations: First, certain concept prompts include shortcuts that recognize correct concepts for wrong reasons; Second, multimodal information (e.g. visual discriminativeness, and textual knowledge) should be leveraged when selecting the concepts. Our proposed concept discovery and learning (CDL) framework is thus designed to identify a diverse list of generic visual concepts (e.g. "spiky" as opposed to "spiky durian"), which are ranked and selected based on visual and language mutual information. We carefully design quantitative and human evaluations of the discovered concepts on six diverse visual recognition datasets, which confirm that pre-trained VLMs do learn visual concepts that provide accurate and thorough descriptions for the recognized objects. All code and models are publicly released.
Keyword: language-vision
There is no result
Keyword: phrase-grounding
There is no result
Keyword: phrase grounding
There is no result
Keyword: reference expression comprehension
There is no result
Keyword: chest
There is no result
Keyword: x-ray
There is no result
Keyword: clinical
CT-ADE: An Evaluation Benchmark for Adverse Drug Event Prediction from Clinical Trial Results
Authors: Anthony Yazdani, Alban Bornet, Boya Zhang, Philipp Khlebnikov, Poorya Amini, Douglas Teodoro
Abstract
Adverse drug events (ADEs) significantly impact clinical research and public health, contributing to failures in clinical trials and leading to increased healthcare costs. The accurate prediction and management of ADEs are crucial for improving the development of safer, more effective medications, and enhancing patient outcomes. To support this effort, we introduce CT-ADE, a novel dataset compiled to enhance the predictive modeling of ADEs. Encompassing over 12,000 instances extracted from clinical trial results, the CT-ADE dataset integrates drug, patient population, and contextual information for multilabel ADE classification tasks in monopharmacy treatments, providing a comprehensive resource for developing advanced predictive models. To mirror the complex nature of ADEs, annotations are standardized at the system organ class level of the Medical Dictionary for Regulatory Activities (MedDRA) ontology. Preliminary analyses using baseline models have demonstrated promising results, achieving 73.33% F1 score and 81.54% balanced accuracy, highlighting CT-ADE's potential to advance ADE prediction. CT-ADE provides an essential tool for researchers aiming to leverage the power of artificial intelligence and machine learning to enhance patient safety and minimize the impact of ADEs on pharmaceutical research and development. Researchers interested in using the CT-ADE dataset can find all necessary resources at https://github.com/xxxx/xxxx.
LaPA: Latent Prompt Assist Model For Medical Visual Question Answering
Authors: Tiancheng Gu, Kaicheng Yang, Dongnan Liu, Weidong Cai
Subjects: Computer Vision and Pattern Recognition (cs.CV); Computation and Language (cs.CL)
Abstract
Medical visual question answering (Med-VQA) aims to automate the prediction of correct answers for medical images and questions, thereby assisting physicians in reducing repetitive tasks and alleviating their workload. Existing approaches primarily focus on pre-training models using additional and comprehensive datasets, followed by fine-tuning to enhance performance in downstream tasks. However, there is also significant value in exploring existing models to extract clinically relevant information. In this paper, we propose the Latent Prompt Assist model (LaPA) for medical visual question answering. Firstly, we design a latent prompt generation module to generate the latent prompt with the constraint of the target answer. Subsequently, we propose a multi-modal fusion block with latent prompt fusion module that utilizes the latent prompt to extract clinical-relevant information from uni-modal and multi-modal features. Additionally, we introduce a prior knowledge fusion module to integrate the relationship between diseases and organs with the clinical-relevant information. Finally, we combine the final integrated information with image-language cross-modal information to predict the final answers. Experimental results on three publicly available Med-VQA datasets demonstrate that LaPA outperforms the state-of-the-art model ARL, achieving improvements of 1.83%, 0.63%, and 1.80% on VQA-RAD, SLAKE, and VQA-2019, respectively. The code is publicly available at https://github.com/GaryGuTC/LaPA_model.
Keyword: biomedical
There is no result
Keyword: radiology
There is no result
Keyword: radiography
There is no result
Keyword: medical
Unlocking Multi-View Insights in Knowledge-Dense Retrieval-Augmented Generation
Abstract
While Retrieval-Augmented Generation (RAG) plays a crucial role in the application of Large Language Models (LLMs), existing retrieval methods in knowledge-dense domains like law and medicine still suffer from a lack of multi-perspective views, which are essential for improving interpretability and reliability. Previous research on multi-view retrieval often focused solely on different semantic forms of queries, neglecting the expression of specific domain knowledge perspectives. This paper introduces a novel multi-view RAG framework, MVRAG, tailored for knowledge-dense domains that utilizes intention-aware query rewriting from multiple domain viewpoints to enhance retrieval precision, thereby improving the effectiveness of the final inference. Experiments conducted on legal and medical case retrieval demonstrate significant improvements in recall and precision rates with our framework. Our multi-perspective retrieval approach unleashes the potential of multi-view information enhancing RAG tasks, accelerating the further application of LLMs in knowledge-intensive fields.
Keyword: active-learning
There is no result
Keyword: active learning
There is no result
Keyword: chexpert
There is no result
Keyword: vision
Adaptive Memory Replay for Continual Learning
Authors: James Seale Smith, Lazar Valkov, Shaunak Halbe, Vyshnavi Gutta, Rogerio Feris, Zsolt Kira, Leonid Karlinsky
Subjects: Machine Learning (cs.LG); Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV)
Abstract
Foundation Models (FMs) have become the hallmark of modern AI, however, these models are trained on massive data, leading to financially expensive training. Updating FMs as new data becomes available is important, however, can lead to `catastrophic forgetting', where models underperform on tasks related to data sub-populations observed too long ago. This continual learning (CL) phenomenon has been extensively studied, but primarily in a setting where only a small amount of past data can be stored. We advocate for the paradigm where memory is abundant, allowing us to keep all previous data, but computational resources are limited. In this setting, traditional replay-based CL approaches are outperformed by a simple baseline which replays past data selected uniformly at random, indicating that this setting necessitates a new approach. We address this by introducing a framework of adaptive memory replay for continual learning, where sampling of past data is phrased as a multi-armed bandit problem. We utilize Bolzmann sampling to derive a method which dynamically selects past data for training conditioned on the current task, assuming full data access and emphasizing training efficiency. Through extensive evaluations on both vision and language pre-training tasks, we demonstrate the effectiveness of our approach, which maintains high performance while reducing forgetting by up to 10% at no training efficiency cost.
Auto-Formula: Recommend Formulas in Spreadsheets using Contrastive Learning for Table Representations
Authors: Sibei Chen, Yeye He, Weiwei Cui, Ju Fan, Song Ge, Haidong Zhang, Dongmei Zhang, Surajit Chaudhuri
Subjects: Databases (cs.DB); Computation and Language (cs.CL); Programming Languages (cs.PL)
Abstract
Spreadsheets are widely recognized as the most popular end-user programming tools, which blend the power of formula-based computation, with an intuitive table-based interface. Today, spreadsheets are used by billions of users to manipulate tables, most of whom are neither database experts nor professional programmers. Despite the success of spreadsheets, authoring complex formulas remains challenging, as non-technical users need to look up and understand non-trivial formula syntax. To address this pain point, we leverage the observation that there is often an abundance of similar-looking spreadsheets in the same organization, which not only have similar data, but also share similar computation logic encoded as formulas. We develop an Auto-Formula system that can accurately predict formulas that users want to author in a target spreadsheet cell, by learning and adapting formulas that already exist in similar spreadsheets, using contrastive-learning techniques inspired by "similar-face recognition" from compute vision. Extensive evaluations on over 2K test formulas extracted from real enterprise spreadsheets show the effectiveness of Auto-Formula over alternatives. Our benchmark data is available at https://github.com/microsoft/Auto-Formula to facilitate future research.
PDF-MVQA: A Dataset for Multimodal Information Retrieval in PDF-based Visual Question Answering
Abstract
Document Question Answering (QA) presents a challenge in understanding visually-rich documents (VRD), particularly those dominated by lengthy textual content like research journal articles. Existing studies primarily focus on real-world documents with sparse text, while challenges persist in comprehending the hierarchical semantic relations among multiple pages to locate multimodal components. To address this gap, we propose PDF-MVQA, which is tailored for research journal articles, encompassing multiple pages and multimodal information retrieval. Unlike traditional machine reading comprehension (MRC) tasks, our approach aims to retrieve entire paragraphs containing answers or visually rich document entities like tables and figures. Our contributions include the introduction of a comprehensive PDF Document VQA dataset, allowing the examination of semantically hierarchical layout structures in text-dominant documents. We also present new VRD-QA frameworks designed to grasp textual contents and relations among document layouts simultaneously, extending page-level understanding to the entire multi-page document. Through this work, we aim to enhance the capabilities of existing vision-and-language models in handling challenges posed by text-dominant documents in VRD-QA.
Keyword: visual
How Does the Textual Information Affect the Retrieval of Multimodal In-Context Learning?
Authors: Yang Luo, Zangwei Zheng, Zirui Zhu, Yang You
Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
Abstract
The increase in parameter size of multimodal large language models (MLLMs) introduces significant capabilities, particularly in-context learning, where MLLMs enhance task performance without updating pre-trained parameters. This effectiveness, however, hinges on the appropriate selection of in-context examples, a process that is currently biased towards visual data, overlooking textual information. Furthermore, the area of supervised retrievers for MLLMs, crucial for optimal in-context example selection, continues to be uninvestigated. Our study offers an in-depth evaluation of the impact of textual information on the unsupervised selection of in-context examples in multimodal contexts, uncovering a notable sensitivity of retriever performance to the employed modalities. Responding to this, we introduce a novel supervised MLLM-retriever MSIER that employs a neural network to select examples that enhance multimodal in-context learning efficiency. This approach is validated through extensive testing across three distinct tasks, demonstrating the method's effectiveness. Additionally, we investigate the influence of modalities on our supervised retrieval method's training and pinpoint factors contributing to our model's success. This exploration paves the way for future advancements, highlighting the potential for refined in-context learning in MLLMs through the strategic use of multimodal data.
Groma: Localized Visual Tokenization for Grounding Multimodal Large Language Models
Abstract
We introduce Groma, a Multimodal Large Language Model (MLLM) with grounded and fine-grained visual perception ability. Beyond holistic image understanding, Groma is adept at region-level tasks such as region captioning and visual grounding. Such capabilities are built upon a localized visual tokenization mechanism, where an image input is decomposed into regions of interest and subsequently encoded into region tokens. By integrating region tokens into user instructions and model responses, we seamlessly enable Groma to understand user-specified region inputs and ground its textual output to images. Besides, to enhance the grounded chat ability of Groma, we curate a visually grounded instruction dataset by leveraging the powerful GPT-4V and visual prompting techniques. Compared with MLLMs that rely on the language model or external module for localization, Groma consistently demonstrates superior performances in standard referring and grounding benchmarks, highlighting the advantages of embedding localization into image tokenization. Project page: https://groma-mllm.github.io/.
Keyword: visio-linguistic
There is no result
Keyword: cross-modal
Cooperative Sentiment Agents for Multimodal Sentiment Analysis
Authors: Shanmin Wang, Hui Shuai, Qingshan Liu, Fei Wang
Subjects: Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV)
Abstract
In this paper, we propose a new Multimodal Representation Learning (MRL) method for Multimodal Sentiment Analysis (MSA), which facilitates the adaptive interaction between modalities through Cooperative Sentiment Agents, named Co-SA. Co-SA comprises two critical components: the Sentiment Agents Establishment (SAE) phase and the Sentiment Agents Cooperation (SAC) phase. During the SAE phase, each sentiment agent deals with an unimodal signal and highlights explicit dynamic sentiment variations within the modality via the Modality-Sentiment Disentanglement (MSD) and Deep Phase Space Reconstruction (DPSR) modules. Subsequently, in the SAC phase, Co-SA meticulously designs task-specific interaction mechanisms for sentiment agents so that coordinating multimodal signals to learn the joint representation. Specifically, Co-SA equips an independent policy model for each sentiment agent that captures significant properties within the modality. These policies are optimized mutually through the unified reward adaptive to downstream tasks. Benefitting from the rewarding mechanism, Co-SA transcends the limitation of pre-defined fusion modes and adaptively captures unimodal properties for MRL in the multimodal interaction setting. To demonstrate the effectiveness of Co-SA, we apply it to address Multimodal Sentiment Analysis (MSA) and Multimodal Emotion Recognition (MER) tasks. Our comprehensive experimental results demonstrate that Co-SA excels at discovering diverse cross-modal features, encompassing both common and complementary aspects. The code can be available at https://github.com/smwanghhh/Co-SA.
Keyword: alignment
mOthello: When Do Cross-Lingual Representation Alignment and Cross-Lingual Transfer Emerge in Multilingual Models?
Beyond Human Norms: Unveiling Unique Values of Large Language Models through Interdisciplinary Approaches
Keyword: aligning
Rethinking the Evaluation of Dialogue Systems: Effects of User Feedback on Crowdworkers and LLMs
Keyword: align
BIRD: A Trustworthy Bayesian Inference Framework for Large Language Models
Data Alignment for Zero-Shot Concept Generation in Dermatology AI
Keyword: vision language
There is no result
Keyword: vision-language
Pre-trained Vision-Language Models Learn Discoverable Visual Concepts
Keyword: language-vision
There is no result
Keyword: phrase-grounding
There is no result
Keyword: phrase grounding
There is no result
Keyword: reference expression comprehension
There is no result
Keyword: chest
There is no result
Keyword: x-ray
There is no result
Keyword: clinical
CT-ADE: An Evaluation Benchmark for Adverse Drug Event Prediction from Clinical Trial Results
LaPA: Latent Prompt Assist Model For Medical Visual Question Answering
Keyword: biomedical
There is no result
Keyword: radiology
There is no result
Keyword: radiography
There is no result
Keyword: medical
Unlocking Multi-View Insights in Knowledge-Dense Retrieval-Augmented Generation
Keyword: active-learning
There is no result
Keyword: active learning
There is no result
Keyword: chexpert
There is no result
Keyword: vision
Adaptive Memory Replay for Continual Learning
Auto-Formula: Recommend Formulas in Spreadsheets using Contrastive Learning for Table Representations
PDF-MVQA: A Dataset for Multimodal Information Retrieval in PDF-based Visual Question Answering
Keyword: visual
How Does the Textual Information Affect the Retrieval of Multimodal In-Context Learning?
Groma: Localized Visual Tokenization for Grounding Multimodal Large Language Models
Keyword: visio-linguistic
There is no result
Keyword: cross-modal
Cooperative Sentiment Agents for Multimodal Sentiment Analysis
Keyword: modality
There is no result
Keyword: modalities
There is no result
Keyword: multi-modal
There is no result
Keyword: multimodal
There is no result