Video Moment Retrieval (VMR) requires precise modelling of fine-grained moment-text associations to capture intricate visual-language relationships. Due to the lack of a diverse and generalisable VMR dataset to facilitate learning scalable moment-text associations, existing methods resort to joint training on both source and target domain videos for cross-domain applications. Meanwhile, recent developments in vision-language multimodal models pre-trained on large-scale image-text and/or video-text pairs are only based on coarse associations (weakly labelled). They are inadequate to provide fine-grained moment-text correlations required for cross-domain VMR. In this work, we solve the problem of unseen cross-domain VMR, where certain visual and textual concepts do not overlap across domains, by only utilising target domain sentences (text prompts) without accessing their videos. To that end, we explore generative video diffusion for fine-grained editing of source videos controlled by the target sentences, enabling us to simulate target domain videos. We address two problems in video editing for optimising unseen domain VMR: (1) generation of high-quality simulation videos of different moments with subtle distinctions, (2) selection of simulation videos that complement existing source training videos without introducing harmful noise or unnecessary repetitions. On the first problem, we formulate a two-stage video diffusion generation controlled simultaneously by (1) the original video structure of a source video, (2) subject specifics, and (3) a target sentence prompt. This ensures fine-grained variations between video moments. On the second problem, we introduce a hybrid selection mechanism that combines two quantitative metrics for noise filtering and one qualitative metric for leveraging VMR prediction on simulation video selection.
No two authors write alike. Personal flourishes invoked in written narratives, from lexicon to rhetorical devices, imply a particular author--what literary theorists label the implied or virtual author; distinct from the real author or narrator of a text. Early large language models trained on unfiltered training sets drawn from a variety of discordant sources yielded incoherent personalities, problematic for conversational tasks but proving useful for sampling literature from multiple perspectives. Successes in alignment research in recent years have allowed researchers to impose subjectively consistent personae on language models via instruction tuning and reinforcement learning from human feedback (RLHF), but whether aligned models retain the ability to model an arbitrary virtual author has received little scrutiny. By studying 4,374 stories sampled from three OpenAI language models, we show successive versions of GPT-3 suffer from increasing degrees of "mode collapse" whereby overfitting the model during alignment constrains it from generalizing over authorship: models suffering from mode collapse become unable to assume a multiplicity of perspectives. Our method and results are significant for researchers seeking to employ language models in sociological simulations.
Image-to-video (I2V) generation aims to use the initial frame (alongside a text prompt) to create a video sequence. A grand challenge in I2V generation is to maintain visual consistency throughout the video: existing methods often struggle to preserve the integrity of the subject, background, and style from the first frame, as well as ensure a fluid and logical progression within the video narrative. To mitigate these issues, we propose ConsistI2V, a diffusion-based method to enhance visual consistency for I2V generation. Specifically, we introduce (1) spatiotemporal attention over the first frame to maintain spatial and motion consistency, (2) noise initialization from the low-frequency band of the first frame to enhance layout consistency. These two approaches enable ConsistI2V to generate highly consistent videos. We also extend the proposed approaches to show their potential to improve consistency in auto-regressive long video generation and camera motion control. To verify the effectiveness of our method, we propose I2V-Bench, a comprehensive evaluation benchmark for I2V generation. Our automatic and human evaluation results demonstrate the superiority of ConsistI2V over existing methods.
Humans make numerous inferences in text comprehension to understand discourse. This paper aims to understand the commonalities and disparities in the inference judgments between humans and state-of-the-art Large Language Models (LLMs). Leveraging a comprehensively curated entailment verification benchmark, we evaluate both human and LLM performance across various reasoning categories. Our benchmark includes datasets from three categories (NLI, contextual QA, and rationales) that include multi-sentence premises and different knowledge types, thereby evaluating the inference capabilities in complex reasoning instances. Notably, our findings reveal LLMs' superiority in multi-hop reasoning across extended contexts, while humans excel in tasks necessitating simple deductive reasoning. Leveraging these insights, we introduce a fine-tuned Flan-T5 model that outperforms GPT-3.5 and rivals with GPT-4, offering a robust open-source solution for entailment verification. As a practical application, we showcase the efficacy of our finetuned model in enhancing self-consistency in model-generated explanations, resulting in a 6% performance boost on average across three multiple-choice question-answering datasets.
Scaling up contrastive language-image pretraining (CLIP) is critical for empowering both vision and multimodal models. We present EVA-CLIP-18B, the largest and most powerful open-source CLIP model to date, with 18-billion parameters. With only 6-billion training samples seen, EVA-CLIP-18B achieves an exceptional 80.7% zero-shot top-1 accuracy averaged across 27 widely recognized image classification benchmarks, outperforming its forerunner EVA-CLIP (5-billion parameters) and other open-source CLIP models by a large margin. Remarkably, we observe a consistent performance improvement with the model size scaling of EVA-CLIP, despite maintaining a constant training dataset of 2-billion image-text pairs from LAION-2B and COYO-700M. This dataset is openly available and much smaller than the in-house datasets (e.g., DFN-5B, WebLI-10B) employed in other state-of-the-art CLIP models. EVA-CLIP-18B demonstrates the potential of EVA-style weak-to-strong visual model scaling. With our model weights made publicly available, we hope to facilitate future research in vision and multimodal foundation models.
Sentence-level attacks craft adversarial sentences that are synonymous with correctly-classified sentences but are misclassified by the text classifiers. Under the black-box setting, classifiers are only accessible through their feedback to queried inputs, which is predominately available in the form of class probabilities. Even though utilizing class probabilities results in stronger attacks, due to the challenges of using them for sentence-level attacks, existing attacks use either no feedback or only the class labels. Overcoming the challenges, we develop a novel algorithm that uses class probabilities for black-box sentence-level attacks, investigate the effectiveness of using class probabilities on the attack's success, and examine the question if it is worthy or practical to use class probabilities by black-box sentence-level attacks. We conduct extensive evaluations of the proposed attack comparing with the baselines across various classifiers and benchmark datasets.
Cued Speech (CS) is a pure visual coding method used by hearing-impaired people that combines lip reading with several specific hand shapes to make the spoken language visible. Automatic CS recognition (ACSR) seeks to transcribe visual cues of speech into text, which can help hearing-impaired people to communicate effectively. The visual information of CS contains lip reading and hand cueing, thus the fusion of them plays an important role in ACSR. However, most previous fusion methods struggle to capture the global dependency present in long sequence inputs of multi-modal CS data. As a result, these methods generally fail to learn the effective cross-modal relationships that contribute to the fusion. Recently, attention-based transformers have been a prevalent idea for capturing the global dependency over the long sequence in multi-modal fusion, but existing multi-modal fusion transformers suffer from both poor recognition accuracy and inefficient computation for the ACSR task. To address these problems, we develop a novel computation and parameter efficient multi-modal fusion transformer by proposing a novel Token-Importance-Aware Attention mechanism (TIAA), where a token utilization rate (TUR) is formulated to select the important tokens from the multi-modal streams. More precisely, TIAA firstly models the modality-specific fine-grained temporal dependencies over all tokens of each modality, and then learns the efficient cross-modal interaction for the modality-shared coarse-grained temporal dependencies over the important tokens of different modalities. Besides, a light-weight gated hidden projection is designed to control the feature flows of TIAA. The resulting model, named Economical Cued Speech Fusion Transformer (EcoCued), achieves state-of-the-art performance on all existing CS datasets, compared with existing transformer-based fusion methods and ACSR fusion methods.
In this paper, we propose a controllable dense captioner (ControlCap), which accommodates user's intention to dense captioning by introducing linguistic guidance. ControlCap is defined as a multimodal embedding bridging architecture, which comprises multimodal embedding generation (MEG) module and bi-directional embedding bridging (BEB) module. While MEG module represents objects/regions by combining embeddings of detailed information with context-aware ones, it also endows ControlCap the adaptability to specialized controls by utilizing them as linguistic guidance. BEB module aligns the linguistic guidance with visual embeddings through borrowing/returning features from/to the visual domain and gathering such features to predict text descriptions. Experiments on Visual Genome and VG-COCO datasets show that ControlCap respectively outperforms the state-of-the-art methods by 1.5% and 3.7% (mAP). Last but not least, with the capability of converting region-category pairs to region-text pairs, ControlCap is able to act as a powerful data engine for dense captioning. Code is available at https://github.com/callsys/ControlCap.
Quantifying uncertainty in automatically generated text is important for letting humans check potential hallucinations and making systems more reliable. Conformal prediction is an attractive framework to provide predictions imbued with statistical guarantees, however, its application to text generation is challenging since any i.i.d. assumptions are not realistic. In this paper, we bridge this gap by leveraging recent results on non-exchangeable conformal prediction, which still ensures bounds on coverage. The result, non-exchangeable conformal nucleus sampling, is a novel extension of the conformal prediction framework to generation based on nearest neighbors. Our method can be used post-hoc for an arbitrary model without extra training and supplies token-level, calibrated prediction sets equipped with statistical guarantees. Experiments in machine translation and language modeling show encouraging results in generation quality. By also producing tighter prediction sets with good coverage, we thus give a more theoretically principled way to perform sampling with conformal guarantees.
The intricate relationship between human decision-making and emotions, particularly guilt and regret, has significant implications on behavior and well-being. Yet, these emotions subtle distinctions and interplay are often overlooked in computational models. This paper introduces a dataset tailored to dissect the relationship between guilt and regret and their unique textual markers, filling a notable gap in affective computing research. Our approach treats guilt and regret recognition as a binary classification task and employs three machine learning and six transformer-based deep learning techniques to benchmark the newly created dataset. The study further implements innovative reasoning methods like chain-of-thought and tree-of-thought to assess the models interpretive logic. The results indicate a clear performance edge for transformer-based models, achieving a 90.4% macro F1 score compared to the 85.3% scored by the best machine learning classifier, demonstrating their superior capability in distinguishing complex emotional states.