21 citations · 27 across the 14 of their papers we have counts for
16 papers
DKL: Decoupled Knowledge Learning for Instruction-Tuned Language Models
Kushagra Bhushan, Meghanadh Pulivarthi, Sai Krishna Reddy Sathi +7
RAG has become the de facto method for incorporating new, corpus-specific knowledge into an instruction following LLM (Instruct LLM). Although RAG-based prompting improves factual…
KItCAT: Knowledge Injection via Input Corruption for Auto-regressive Training
Meghanadh Pulivarthi, Kushagra Bhushan, Vineet Kumar +5
LLMs acquire vast amounts of knowledge during pre-training, but often lack the specialized knowledge needed to answer questions from niche sources such as manuals or technical docu…
ToolWeave: Structured Synthesis of Complex Multi-Turn Tool-Calling Dialogues
Dinesh Khandelwal, Gnana Prakash Punnavajhala, GPS Bhargav +4
Multi-turn tool calling is essential for LLMs to function as autonomous agents, yet synthesizing the training data required for these capabilities remains a fundamental challenge.…
Insertion Based Sequence Generation with Learnable Order Dynamics
Dhruvesh Patel, Benjamin Rozonoyer, Gaurav Pandey +3
Existing insertion-based masked diffusion models that generate sequences by interleaving token insertion with unmasking use fixed schedules that are not dependent on the data. For…
Selective Self-Rehearsal: A Fine-Tuning Approach to Improve Generalization in Large Language Models
Sonam Gupta, Yatin Nandwani, Asaf Yehudai +4
Fine-tuning Large Language Models (LLMs) on specific datasets is a common practice to improve performance on target tasks. However, this performance gain often leads to overfitting…
BRAIn: Bayesian Reward-conditioned Amortized Inference for natural language generation from feedback
Gaurav Pandey, Yatin Nandwani, Tahira Naseem +6
Distribution matching methods for language model alignment such as Generation with Distributional Control (GDC) and Distributional Policy Gradient (DPG) have not received the same…
Mix-and-Match: Scalable Dialog Response Retrieval using Gaussian Mixture Embeddings
Gaurav Pandey, Danish Contractor, Sachindra Joshi
Embedding-based approaches for dialog response retrieval embed the context-response pairs as points in the embedding space. These approaches are scalable, but fail to account for t…
Gaining Insights into Unrecognized User Utterances in Task-Oriented Dialog Systems
Ella Rabinovich, Matan Vetzler, David Boaz +3
The rapidly growing market demand for automatic dialogue agents capable of goal-oriented behavior has caused many tech-industry leaders to invest considerable efforts into task-ori…
Variational Learning for Unsupervised Knowledge Grounded Dialogs
Mayank Mishra, Dhiraj Madan, Gaurav Pandey +1
Recent methods for knowledge grounded dialogs generate responses by incorporating information from an external textual document. These methods do not require the exact document to…
Simulated Chats for Building Dialog Systems: Learning to Generate Conversations from Instructions
Biswesh Mohapatra, Gaurav Pandey, Danish Contractor +1
Popular dialog datasets such as MultiWOZ are created by providing crowd workers an instruction, expressed in natural language, that describes the task to be accomplished. Crowd wor…
Unsupervised feature learning with discriminative encoder
Gaurav Pandey, Ambedkar Dukkipati
In recent years, deep discriminative models have achieved extraordinary performance on supervised learning tasks, significantly outperforming their generative counterparts. However…
Discriminative Neural Topic Models
Gaurav Pandey, Ambedkar Dukkipati
We propose a neural network based approach for learning topics from text and image datasets. The model makes no assumptions about the conditional distribution of the observed featu…
On collapsed representation of hierarchical Completely Random Measures
Gaurav Pandey, Ambedkar Dukkipati
The aim of the paper is to provide an exact approach for generating a Poisson process sampled from a hierarchical CRM, without having to instantiate the infinitely many atoms of th…
To go deep or wide in learning?
Gaurav Pandey, Ambedkar Dukkipati
To achieve acceptable performance for AI tasks, one can either use sophisticated feature extraction methods as the first layer in a two-layered supervised learning model, or learn…
Generative Maximum Entropy Learning for Multiclass Classification
Ambedkar Dukkipati, Gaurav Pandey, Debarghya Ghoshdastidar +2
Maximum entropy approach to classification is very well studied in applied statistics and machine learning and almost all the methods that exists in literature are discriminative i…
Minimum Description Length Principle for Maximum Entropy Model Selection
Gaurav Pandey, Ambedkar Dukkipati
Model selection is central to statistics, and many learning problems can be formulated as model selection problems. In this paper, we treat the problem of selecting a maximum entro…