Krishna Sayana
Krishna Sayana is a Software Engineer at Google Research working on personalized conversational recommender models and reinforcement learning frameworks for prompt and agent optimization across Search, Shopping, and Ads.
Authored Publications
Sort By
Beyond Retrieval: Generating Narratives in Conversational Recommender Systems
Raghavendra Vasudeva
Yuri Vasilevski
Kun Su
Liam Hebert
James Pine
Hubert Pham
Ambarish Jash
Sukhdeep Sodhi
(2025)
Preview abstract
Large Language Models (LLMs) have shown remarkable progress in generating human-quality text and engaging in complex reasoning. This presents a unique opportunity to revolutionize conversational recommender systems by enabling them to generate rich, engaging and personalized narratives that go beyond recommendations. However, the lack of suitable datasets limits research in this area. This paper addresses this challenge by making two key contributions.
First, we introduce REGEN Reviews Enhanced with GEnerative Narratives, a new dataset extending the Amazon Product Reviews with rich user narratives. Furthermore, we perform an extensive automated evaluation of the dataset using a rater LLM. Second, the paper introduces a fusion architecture (CF model with an LLM) which serves as a baseline for REGEN. To the best of our knowledge, this represents the first attempt to analyze the capabilities of LLMs in understanding recommender signals and generating rich narratives. We demonstrate that LLMs can effectively learn from simple fusion architectures utilizing interaction-based CF embeddings, and this can be further enhanced using the metadata and personalization data associated with items. Our experiments show that combining CF and content embeddings leads to improvements of 4-12% in key language metrics compared to using either type of embedding individually. We also provide an analysis to interpret their contributions to this new generative task.
View details
Preview abstract
Integrating product catalogs and user behavior into LLMs can enhance recommendations with broad world knowledge, but the scale of real-world item catalogs, often containing millions of discrete item identifiers (Item IDs), poses a significant challenge. This contrasts with the smaller, tokenized text vocabularies typically used in LLMs. The predominant view within the LLM-based recommendation literature is that it is infeasible to treat item ids as a first class citizen in the LLM and instead some sort of tokenization of an item into multiple tokens is required. However, this creates a key practical bottleneck in serving these models for real-time low-latency applications.
Our paper challenges this predominant practice and integrates item ids as first class citizens into the LLM. We provide simple, yet highly effective, novel training and inference modifications that enable single-token representations of items and single-step decoding. Our method shows improvements in recommendation quality (Recall and NDCG) over existing techniques on the Amazon shopping datasets while significantly improving inference efficiency by 5x-14x. Our work offers an efficiency perspective distinct from that of other popular approaches within LLM-based recommendation, potentially inspiring further research and opening up a new direction for integrating IDs into LLMs. Our code is available here https://drive.google.com/file/d/1cUMj37rV0Z1bCWMdhQ6i4q4eTRQLURtC/edit
View details
REGEN: A Dataset and Benchmarks with Natural Language Critiques and Narratives
Kun Su
Hubert Pham
James Pine
Yuri Vasilevski
Raghavendra Vasudeva
Liam Hebert
Ambarish Jash
Anushya Subbiah
Sukhdeep Sodhi
(2025)
Preview abstract
This paper introduces a novel dataset REGEN (Reviews Enhanced with GEnerative Narratives), designed to benchmark the conversational capabilities of recommender Large Language Models (LLMs), addressing the limitations of existing datasets that primarily focus on sequential item prediction. REGEN extends the Amazon Product Reviews dataset by inpainting two key natural language features: (1) user critiques, representing user "steering" queries that lead to the selection of a subsequent item, and (2) narratives, rich textual outputs associated with each recommended item taking into account prior context. The narratives include product endorsements, purchase explanations, and summaries of user preferences.
Further, we establish an end-to-end modeling benchmark for the task of conversational recommendation, where models are trained to generate both recommendations and corresponding narratives conditioned on user history (items and critiques). For this joint task, we introduce a modeling framework LUMEN (LLM-based Unified Multi-task Model with Critiques, Recommendations, and Narratives) which uses an LLM as a backbone for critiquing, retrieval and generation. We also evaluate the dataset's quality using standard auto-rating techniques and benchmark it by training both traditional and LLM-based recommender models. Our results demonstrate that incorporating critiques enhances recommendation quality by enabling the recommender to learn language understanding and integrate it with recommendation signals. Furthermore, LLMs trained on our dataset effectively generate both recommendations and contextual narratives, achieving performance comparable to state-of-the-art recommenders and language models.
View details
PERSOMA: PERsonalized SOft ProMpt Adapter Architecture for Personalized Language Prompting
Liam Hebert
Ambarish Jash
Alexandros Karatzoglou
Sukhdeep Sodhi
Sumanth Doddapaneni
Yanli Cai
Dima Kuzmin
2024
Preview abstract
Understanding extensive historical user interactions is pivotal in
capturing users’ evolving preferences, enabling more precise and
personalized natural language systems. To tackle this challenge,
we introduce the PERSOMA: Personalized Soft Prompt Adapter
architecture. In contrast to previous work in personalized prompt-
ing using large language models, PERSOMA introduces a novel
approach to efficiently capture user history in free-form text by re-
sampling and compressing interactions as expressive soft prompt
embeddings. We validate our approach through an extensive evalua-
tion of various adapter architectures, first stage sampling strategies,
and other personalization methods. Our results demonstrate the
superior capability of PERSOMA in handling large complex histories
compared to previous embedding-based and text-prompt based methods.
View details
Preview abstract
Modeling long user histories plays a pivotal role in enhancing recommendation systems, allowing them to capture users' evolving preferences, resulting in more precise and personalized recommendations. In this study, we tackle the challenges of modeling long user histories for preference understanding in natural language. We introduce a new User Embedding Module that efficiently processes user history in free-form text by compressing and representing them as embeddings, and using them as soft prompts to a language model. Our experiments demonstrate the superior capability of this approach in handling significantly longer histories compared to conventional text-based methods, yielding substantial improvements in predictive performance. Models trained using our approach exhibit substantial enhancements, with up to 0.21 and 0.25 F1 points improvement over the text-based prompting baselines. The main contribution of this research is to demonstrate the ability to bias language models via user signals (preferences).
View details
