Papers by Ashish Kumar

14 papers
EGOILLUSION: Benchmarking Hallucinations in Egocentric Video Understanding (2025.emnlp-main)

Copied to clipboard

Challenge: Multimodal Large Language Models excel at visual perception and reasoning in third-person and egocentric videos, but are prone to hallucinations, generating coherent yet inaccurate responses.
Approach: They propose to use a benchmark to evaluate MLLM hallucinations in egocentric videos.
Outcome: EGOILLUSION comprises 1,400 videos paired with 8,000 human-annotated open and closed-ended questions designed to trigger hallucinations in both visual and auditory cues in egocentric videos.
Semantic alignment in hyperbolic space for fine-grained emotion classification (2025.acl-srw)

Copied to clipboard

Challenge: Existing approaches to fine-grained emotion classification operate in Euclidean space, where the flat geometry makes it difficult to distinguish semantically similar label labels.
Approach: They propose a semantic alignment framework that leverages the Lorentz model of hyperbolic space to embed text and label representations into hyperbolical space via the exponential map.
Outcome: The proposed framework improves on two benchmark FEC datasets.
GAMA: A Large Audio-Language Model with Advanced Audio Understanding and Complex Reasoning Abilities (2024.emnlp-main)

Copied to clipboard

Challenge: We propose a novel large-scale audio-language model with advanced audio understanding and reasoning abilities.
Approach: They propose a general-purpose large audio-language model with advanced audio understanding and reasoning abilities that integrates an LLM with multiple types of audio representations.
Outcome: The proposed model outperforms existing models on audio understanding tasks by 1%-84%.
HyILR: Hyperbolic Instance-Specific Local Relationships for Hierarchical Text Classification (2025.acl-srw)

Copied to clipboard

Challenge: Hierarchical text classification models rely on capturing global label hierarchy, which contains static and redundant relationships.
Approach: They propose a method which captures hierarchical relationships without encoding global hierarchy . they use hyperbolic geometry to model instance-specific local relationships using Lorentz model .
Outcome: The proposed model captures hierarchical relationships without encoding global hierarchy . the proposed model is superior to baseline methods on four benchmark datasets .
CEBC: Conformal Evidence-Bounded Control for Low-Hallucination Vision–Language Generation (2026.acl-long)

Copied to clipboard

Challenge: Existing mitigation approaches reduce hallucinated object mentions at the cost of degraded generation quality or require expensive retraining and task-specific supervision.
Approach: They propose a lightweight framework for low-hallucination vision–language generation . it uses evidence-bounded minimal editing to revise or suppress unsupported referenced entities .
Outcome: The proposed framework reduces hallucinations while maintaining or improving quality metrics.
PAT: Parameter-Free Audio-Text Aligner to Boost Zero-Shot Audio Classification (2025.naacl-long)

Copied to clipboard

Challenge: Audio-Language Models (ALMs) have demonstrated remarkable performance in zero-shot audio classification.
Approach: They propose a training-free method that enhances audio and language representations using mutual feedback.
Outcome: The proposed method outperforms vanilla zero-shot evaluation with significant margins of 0.42%-27.0%.
How much coffee was consumed during EMNLP 2019? Fermi Problems: A New Reasoning Challenge for AI (2021.emnlp-main)

Copied to clipboard

Challenge: a new reasoning challenge is proposed to help AI systems to solve real-world problems . Fermi Problems are questions whose answers can only be approximated because their computation is either impossible or impossible.
Approach: They propose a new reasoning challenge, Fermi Problems, which asks questions whose answers can only be approximated because their computation is either impractical or impossible.
Outcome: The proposed datasets show that even fine-tuned large-scale language models perform poorly on these datasets.
ODASim: Ordered, Distinctive and Absolute Semantic Similarity for Code Explanation Evaluation (2026.findings-acl)

Copied to clipboard

Challenge: Existing methods for code explanations fail to distinguish correct from partially or fully incorrect explanations and their similarity scores are poorly calibrated.
Approach: They propose a model-agnostic graded fine-tuning framework that learns calibrated similarity representations between code and explanations to support fine-grained supervision and evaluation.
Outcome: The proposed framework improves F1 score and ECE scores on two embedding models and reduces expected calibration error.
TEN: Table Explicitization, Neurosymbolically (2026.acl-industry)

Copied to clipboard

Challenge: Existing methods for extracting tabular data from semistructured text are error-prone and costly.
Approach: They propose a neurosymbolic approach to extract tabular data from semistructured text . TEN is a triadic feedback loop that iteratively refines table hypotheses .
Outcome: The proposed approach outperforms neural baselines in exact match accuracy and lower hallucination rates.
MULTIVOX: A Benchmark for Evaluating Voice Assistants for Multimodal Interactions (2025.emnlp-main)

Copied to clipboard

Challenge: omni models lack spoken dialogues, which is essential for assessing conversational and auditory capabilities of voice assistants.
Approach: They propose a benchmark to evaluate the ability of voice assistants to integrate paralinguistic speech features into their models.
Outcome: The multivox voice assistant benchmark evaluates the ability of models to integrate spoken and visual cues including paralinguistic speech features for truly multimodal understanding.
EH-MAM: Easy-to-Hard Masked Acoustic Modeling for Self-Supervised Speech Representation Learning (2024.emnlp-main)

Copied to clipboard

Challenge: EH-MAM is a self-supervised learning approach for speech representation learning . prior methods used random masking schemes to learn speech representations .
Approach: They propose a self-supervised approach that automatically selects hard regions during SSL training and introduces them to the model for reconstruction.
Outcome: The proposed approach outperforms state-of-the-art models across low-resource speech recognition and SUPERB benchmarks by 5%-10%.
Do Audio-Language Models Understand Linguistic Variations? (2025.naacl-short)

Copied to clipboard

Challenge: Existing open-vocabulary audio language models struggle to generalize to linguistic variations in textual queries.
Approach: They propose a novel technique to learn audio-language representations agnostic to linguistic variations by reformulating contrastive loss used in CLAP architectures.
Outcome: The proposed approach improves the performance of the open-vocabulary audio language models by 0.8%-13% across benchmarks and enhances robustness to linguistic variation.
INDIC QA BENCHMARK: A Multilingual Benchmark to Evaluate Question Answering capability of LLMs for Indic Languages (2025.findings-naacl)

Copied to clipboard

Challenge: Large Language Models perform well on unseen tasks in English, but their abilities in non-English languages are less explored due to limited benchmarks and training data.
Approach: They propose to release a large dataset for context-grounded question answering in 11 major Indian languages.
Outcome: The Indic-QA Benchmark compared large datasets of large LLMs on extractive and abstractive tasks in 11 major Indian languages.
PolyAudio: Advancing Multi-Audio Reasoning in Large Audio Language Models with Interleaved Multi-Audio Contexts (2026.findings-acl)

Copied to clipboard

Challenge: Large Audio Language Models have shown impressive performance on single-clip tasks . however, their ability to reason over interleaved multi-audio contexts remains limited .
Approach: They propose a LALM that targets multi-audio understanding via instruction tuning rather than massive-scale pre-training.
Outcome: The proposed model outperforms baseline models on multi-audio tasks while maintaining robustness.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations