arXiv subjects

Gauri Jagatap

Publications and source records attributed to Gauri Jagatap.

12 recordsLinked to original sources

arXiv2026

Video2Reaction: Training Foundation Video Models to Predict Audience Reaction

We introduce Video2Reaction, a multimodal dataset that maps short movie segments to the induced emotional reactions of viewers in the wild, as expressed through social media comments. Video2Reaction captures the natural diversity of emotional responses by aggregating reactions from online comments at scale, modeling labels as distributions over categorical emotions to better reflect the subjective and ambiguous nature of emotional perception. We benchmark two vision-language models (VLMs) finetuned with LoRA, showing that VLMs learn effectively from Video2Reaction and outperform specialized baselines on dominant reaction prediction. We further demonstrate that VLMs pre-finetuned on Video2Reaction transfer effectively to VCE, another induced emotion dataset with a different taxonomy and video domain. Notably, LLaVA-NeXT-Video-7B pre-finetuned on Video2Reaction and adapted on only 1% of VCE training data achieves a top-3 accuracy of 0.682, on par with the best reported VCE performance trained on the full dataset. The dataset is available at https://huggingface.co/datasets/infofusionlab/Video2Reaction

cs.CV↗

arXiv2026

Video2Reaction: Mapping Video to Audience Reaction Distribution in the Wild

Understanding and forecasting audience reactions to video content are crucial for improving content creation, recommendation systems, and media analysis. To enable audience reaction prediction and other content engagement applications, we introduce $\textbf{Video2Reaction}$, a multimodal dataset that maps short movie segments to a distribution of $\textit{induced emotions}$ of viewers in the wild, as expressed through social media. $\textbf{Video2Reaction}$ spans more than 10,000 videos and serves as a reliable benchmark as well as a training resource for audience reaction prediction. To enable cost-effective continuous annotations as reactions may change over time, we develop a two-stage multi-agent pipeline using only open-source LLMs, achieving 86% correctness under blind human verification despite the inherently noisy and subjective nature of the task. We establish the first benchmark for video-to-reaction-distribution prediction in the wild and show that pretrained foundation video models fail in zero-shot settings, while finetuning transforms them into state-of-the-art predictors capable of modeling both full reaction distributions and dominant responses from video alone. However, the task remains challenging: even the strongest methods achieve only 77% Top-3 F1 in dominant reaction prediction (LLaVA-Next), highlighting a substantial gap in modeling collective audience reaction. \modification{Dataset and code are available at our project page: https://information-fusion-lab-umass.github.io/video2reaction-bench.github.io

cs.CV↗

arXiv2026

OnPoint: Offline-to-Online Multi-Level Distillation for Point-Supervised Online Temporal Action Localization

Temporal Action Localization (TAL) typically relies on segment annotations or offline access to full videos, limiting scalability and online use. We introduce Point-Supervised Online TAL (POTAL), which localizes actions in streaming videos using only one temporal point per instance. To solve POTAL, we propose OnPoint, an offline-to-online multi-level distillation framework that transfers knowledge from a point-supervised offline teacher to an online student via (i) pseudo-segment instance distillation, (ii) class-activation sequence distillation, and (iii) anticipatory window-level distillation. We further improve robustness by incorporating the original point labels into student training and by refining anchor decoding with actionness-guided attention calibration. Experiments on five datasets show OnPoint consistently outperforms strong baselines, establishing a solid foundation for POTAL.

cs.CV↗

arXiv2025

Towards Mitigating Hallucinations in Large Vision-Language Models by Refining Textual Embeddings

Hallucinations in Large Vision-Language Models (LVLMs) remain a persistent challenge, often stemming from inadequate integration of visual information during multimodal reasoning. A key cause is the model's over-reliance on textual priors and underutilization of visual cues, leading to outputs that are linguistically fluent but visually inaccurate. For example, given an image of an empty kitchen countertop, an LVLM might hallucinate a "bowl of fruit" or "cup of coffee", relying on language associations rather than visual evidence. Most LVLMs incorporate visual features by appending them to the input stream of a pre-trained LLM and training on large-scale vision-language datasets. Our systematic analysis reveals that this strategy often leads to over-dependence on textual information due to the inherent bias of LLMs towards language-dominant representations. This imbalance skews attention towards the text over visual content, weakening the model's ability to ground outputs in visual inputs. To address this, we propose a simple yet effective visual feature incorporation method that encourages the model to learn visually-informed textual embeddings distinct from those of the base LLM and promotes a more balanced attention distribution. Experimental results across multiple hallucination benchmarks demonstrate that our method significantly reduces hallucinations and fosters more balanced multimodal reasoning. Notably, our approach achieves substantial gains, including +9.33% on MMVP-MLLM, +2.99% on POPE-AOKVQA, up to +3.4% on Merlin, and +3% on the hard-data split of HallusionBench.

cs.CV↗

arXiv2025

Moment Sampling in Video LLMs for Long-Form Video QA

Recent advancements in video large language models (Video LLMs) have significantly advanced the field of video question answering (VideoQA). While existing methods perform well on short videos, they often struggle with long-range reasoning in longer videos. To scale Video LLMs for longer video content, frame sub-sampling (selecting frames at regular intervals) is commonly used. However, this approach is suboptimal, often leading to the loss of crucial frames or the inclusion of redundant information from multiple similar frames. Missing key frames impairs the model's ability to answer questions accurately, while redundant frames lead the model to focus on irrelevant video segments and increase computational resource consumption. In this paper, we investigate the use of a general-purpose text-to-video moment retrieval model to guide the frame sampling process. We propose "moment sampling", a novel, model-agnostic approach that enables the model to select the most relevant frames according to the context of the question. Specifically, we employ a lightweight moment retrieval model to prioritize frame selection. By focusing on the frames most pertinent to the given question, our method enhances long-form VideoQA performance in Video LLMs. Through extensive experiments on four long-form VideoQA datasets, using four state-of-the-art Video LLMs, we demonstrate the effectiveness of the proposed approach.

cs.CV↗

arXiv2023

Can deepfakes be created by novice users?

Recent advancements in machine learning and computer vision have led to the proliferation of Deepfakes. As technology democratizes over time, there is an increasing fear that novice users can create Deepfakes, to discredit others and undermine public discourse. In this paper, we conduct user studies to understand whether participants with advanced computer skills and varying levels of computer science expertise can create Deepfakes of a person saying a target statement using limited media files. We conduct two studies; in the first study (n = 39) participants try creating a target Deepfake in a constrained time frame using any tool they desire. In the second study (n = 29) participants use pre-specified deep learning-based tools to create the same Deepfake. We find that for the first study, 23.1% of the participants successfully created complete Deepfakes with audio and video, whereas, for the second user study, 58.6% of the participants were successful in stitching target speech to the target video. We further use Deepfake detection software tools as well as human examiner-based analysis, to classify the successfully generated Deepfake outputs as fake, suspicious, or real. The software detector classified 80% of the Deepfakes as fake, whereas the human examiners classified 100% of the videos as fake. We conclude that creating Deepfakes is a simple enough task for a novice user given adequate tools and time; however, the resulting Deepfakes are not sufficiently real-looking and are unable to completely fool detection software as well as human examiners

cs.CR↗

arXiv2021

Adversarial Token Attacks on Vision Transformers

Vision transformers rely on a patch token based self attention mechanism, in contrast to convolutional networks. We investigate fundamental differences between these two families of models, by designing a block sparsity based adversarial token attack. We probe and analyze transformer as well as convolutional models with token attacks of varying patch sizes. We infer that transformer models are more sensitive to token attacks than convolutional models, with ResNets outperforming Transformer models by up to $\sim30\%$ in robust accuracy for single token attacks.

cs.CV↗

arXiv2021

Provable Compressed Sensing with Generative Priors via Langevin Dynamics

Deep generative models have emerged as a powerful class of priors for signals in various inverse problems such as compressed sensing, phase retrieval and super-resolution. Here, we assume an unknown signal to lie in the range of some pre-trained generative model. A popular approach for signal recovery is via gradient descent in the low-dimensional latent space. While gradient descent has achieved good empirical performance, its theoretical behavior is not well understood. In this paper, we introduce the use of stochastic gradient Langevin dynamics (SGLD) for compressed sensing with a generative prior. Under mild assumptions on the generative model, we prove the convergence of SGLD to the true signal. We also demonstrate competitive empirical performance to standard gradient descent.

stat.ML↗

arXiv2021

Adversarially Robust Learning via Entropic Regularization

In this paper we propose a new family of algorithms, ATENT, for training adversarially robust deep neural networks. We formulate a new loss function that is equipped with an additional entropic regularization. Our loss function considers the contribution of adversarial samples that are drawn from a specially designed distribution in the data space that assigns high probability to points with high loss and in the immediate neighborhood of training samples. Our proposed algorithms optimize this loss to seek adversarially robust valleys of the loss landscape. Our approach achieves competitive (or better) performance in terms of robust classification accuracy as compared to several state-of-the-art robust learning approaches on benchmark datasets such as MNIST and CIFAR-10.

cs.LG↗

arXiv2020

Algorithmic Guarantees for Inverse Imaging with Untrained Network Priors

Deep neural networks as image priors have been recently introduced for problems such as denoising, super-resolution and inpainting with promising performance gains over hand-crafted image priors such as sparsity and low-rank. Unlike learned generative priors they do not require any training over large datasets. However, few theoretical guarantees exist in the scope of using untrained neural network priors for inverse imaging problems. We explore new applications and theory for untrained neural network priors. Specifically, we consider the problem of solving linear inverse problems, such as compressive sensing, as well as non-linear problems, such as compressive phase retrieval. We model images to lie in the range of an untrained deep generative network with a fixed seed. We further present a projected gradient descent scheme that can be used for both compressive sensing and phase retrieval and provide rigorous theoretical guarantees for its convergence. We also show both theoretically as well as empirically that with deep network priors, one can achieve better compression rates for the same image quality compared to hand crafted priors.

cs.CV↗

arXiv2018

Learning ReLU Networks via Alternating Minimization

We propose and analyze a new family of algorithms for training neural networks with ReLU activations. Our algorithms are based on the technique of alternating minimization: estimating the activation patterns of each ReLU for all given samples, interleaved with weight updates via a least-squares step. The main focus of our paper are 1-hidden layer networks with $k$ hidden neurons and ReLU activation. We show that under standard distributional assumptions on the $d-$dimensional input data, our algorithm provably recovers the true `ground truth' parameters in a linearly convergent fashion. This holds as long as the weights are sufficiently well initialized; furthermore, our method requires only $n=\widetilde{O}(dk^2)$ samples. We also analyze the special case of 1-hidden layer networks with skipped connections, commonly used in ResNet-type architectures, and propose a novel initialization strategy for the same. For ReLU based ResNet type networks, we provide the first linear convergence guarantee with an end-to-end algorithm. We also extend this framework to deeper networks and empirically demonstrate its convergence to a global minimum.

cs.LG↗

arXiv2017

Sample-Efficient Algorithms for Recovering Structured Signals from Magnitude-Only Measurements

We consider the problem of recovering a signal $\mathbf{x}^* \in \mathbf{R}^n$, from magnitude-only measurements $y_i = |\left\langle\mathbf{a}_i,\mathbf{x}^*\right\rangle|$ for $i=[m]$. Also called the phase retrieval, this is a fundamental challenge in bio-,astronomical imaging and speech processing. The problem above is ill-posed; additional assumptions on the signal and/or the measurements are necessary. In this paper we first study the case where the signal $\mathbf{x}^*$ is $s$-sparse. We develop a novel algorithm that we call Compressive Phase Retrieval with Alternating Minimization, or CoPRAM. Our algorithm is simple; it combines the classical alternating minimization approach for phase retrieval with the CoSaMP algorithm for sparse recovery. Despite its simplicity, we prove that CoPRAM achieves a sample complexity of $O(s^2\log n)$ with Gaussian measurements $\mathbf{a}_i$, matching the best known existing results; moreover, it demonstrates linear convergence in theory and practice. Additionally, it requires no extra tuning parameters other than signal sparsity $s$ and is robust to noise. When the sorted coefficients of the sparse signal exhibit a power law decay, we show that CoPRAM achieves a sample complexity of $O(s\log n)$, which is close to the information-theoretic limit. We also consider the case where the signal $\mathbf{x}^*$ arises from structured sparsity models. We specifically examine the case of block-sparse signals with uniform block size of $b$ and block sparsity $k=s/b$. For this problem, we design a recovery algorithm Block CoPRAM that further reduces the sample complexity to $O(ks\log n)$. For sufficiently large block lengths of $b=Θ(s)$, this bound equates to $O(s\log n)$. To our knowledge, this constitutes the first end-to-end algorithm for phase retrieval where the Gaussian sample complexity has a sub-quadratic dependence on the signal sparsity level.

stat.ML↗