Karthik Raman
Karthik is a Senior Staff Research Scientist at Google Research. Along with the team he leads (Hummingbird), his work revolves around Large Language Models and making them more practically useful and usable. Prior to joining Google, Karthik received his PhD from Cornell University in 2015.
A list of pre-Google publications can be found here: http://www.cs.cornell.edu/~karthik/publications.html
Authored Publications
Sort By
Preview abstract
Sequence labeling is a core task in text understanding for IE/IR systems. Text generation models have increasingly become the go-to solution for such tasks (e.g., entity extraction and dialog slot filling). While most research has focused on the labeling accuracy, a key aspect -- of vital practical importance -- has slipped through the cracks: understanding model confidence. More specifically, we lack a principled understanding of how to reliably gauge the confidence of a model in its predictions for each labeled span. This paper aims to provide some empirical insights on estimating model confidence for generative sequence labeling. Most notably, we find that simply using the decoder's output probabilities is not the best in realizing well-calibrated confidence estimates. As verified over six public datasets of different tasks, we show that our proposed approach -- which leverages statistics from top-k predictions by a beam search -- significantly reduces calibration errors of the predictions of a generative sequence labeling model.
View details
Its All Relative! -- A Synthetic Query Generation Approach for Improving Zero-Shot Relevance Prediction
Findings of the Association for Computational Linguistics: NAACL 2024
Preview abstract
Recent developments in large language models (LLMs) have shown promise in their ability to generate synthetic query-document pairs by prompting LLMs with as few as 8 demonstrations \cite{dai2022promptagator}.
This has enabled building better IR models especially for tasks which have no training data readily available.
Typically, such synthetic query generation (QGen) approaches condition on an input context (e.g. document) and generate a query that is relevant to that context or condition the QGen model additionally on the relevance label (e.g. relevant vs irrelevant) to generate queries across relevance buckets.
However, we find that such QGen approaches are sub-optimal as it requires the model to reason about the desired label and the input from only a handful of examples, which is not trivial, especially when the relevance buckets are nuanced.
In this work, we propose to reduce this burden of LLMs by generating queries simultaneously for different labels (e.g. relevance buckets).
We hypothesize that instead of asking the model to generate, say, an irrelevant query given an input context, asking the model to generate an irrelevant query with respect to a relevant query is a much simpler task setup for the model to reason about.
Extensive experimentation across seven IR datasets shows that synthetic queries generated in such a fashion translates to a better downstream performance, suggesting that the generated queries are indeed of higher quality.
View details
Preview abstract
In-Context Learning (ICL) is an emergent capability of Large Language Models (LLMs).
Only a few demonstrations enable LLMs to be used as blackbox for new tasks. Previous studies have shown that using LLMs' outputs as labels is effective in training models to select demonstrations. Such a label is expected to estimate utility of a demonstration in ICL;
however, it has not been well understood how different labeling strategies affect results on target tasks. This paper presents an analysis on different utility functions by focusing on LLMs' output probability given ground-truth output, and task-specific reward given LLMs' prediction. Unlike the previous work, we introduce a novel labeling method, incremental utility, which estimates how much incremental knowledge is brought into the LLMs by a demonstration. We conduct experiments with instruction-tuned LLMs on binary/multi-class classification, segmentation, and translation across Arabic, English, Finnish, Japanese, and Spanish. Our results show that (1) the probability is effective when the probability values are distributed across the whole value range (on the classification tasks), and (2) the downstream metric is more robust when nuanced reward values are provided with long outputs (on the segmentation and translation tasks). We then show that the proposed incremental utility further helps ICL by contrasting how the LLMs perform with and without the demonstrations.
View details
Preview abstract
Query-document relevance prediction is a critical problem in Information Retrieval systems. This problem has increasingly been tackled using (pretrained) transformer-based models which are finetuned using large collections of labeled data. However, in specialized domains such as e-commerce and healthcare, the viability of this approach is limited by the dearth of large in-domain data. To address this paucity, recent methods leverage these powerful models to generate high-quality task and domain-specific synthetic data. Prior work has largely explored synthetic data generation or query generation (QGen) for Question-Answering (QA) and binary (yes/no) relevance prediction, where for instance, the QGen models are given a document, and trained to generate a query relevant to that document. However in many problems, we have a more fine-grained notion of relevance than a simple yes/no label. Thus, in this work, we conduct
a detailed study into how QGen approaches can be leveraged for nuanced relevance prediction. We demonstrate that – contrary to claims from prior works – current QGen approaches fall short of the more conventional cross-domain transfer-learning approaches. Via empirical studies spanning three public e-commerce benchmarks, we identify new shortcomings of existing QGen approaches – including their inability to distinguish between different grades of
relevance. To address this, we introduce label-conditioned QGen models which incorporates knowledge about the different relevance. While our experiments demonstrate that these modifications help improve performance of QGen techniques, we also find that QGen approaches struggle to capture the full nuance of the relevance label space and as a result the generated queries are not faithful to the desired relevance label.
View details
Multi-Task Retrieval-Augmented Text Generation with Relevance Sampling
Sebastian Hofstätter
Jiecao Chen
Hamed Zamani
ICML 2022 Workshop on Knowledge Retrieval and Language Models
Preview abstract
This paper studies multi-task training of retrieval-augmented generation models for knowledge-intensive tasks. We propose to clean the training set by utilizing a distinct property of knowledge-intensive generation: The connection of query-answer pairs to items in the knowledge base. We filter training examples via a threshold of confidence on the relevance labels, whether a pair is answerable by the knowledge base or not.We train a single Fusion-in-Decoder (FiD) generator on seven combined tasks of the KILT benchmark. The experimental results suggest that our simple yet effective approach substantially improves competitive baselines on two strongly imbalanced tasks; and shows either smaller improvements or no significant regression on the remaining tasks. Furthermore, we demonstrate our multi-task training with relevance label sampling scales well with increased model capacity and achieves state-of-the-art results in five out of seven KILT tasks.
View details
FiD-Light: Efficient and Effective Retrieval-Augmented Text Generation
Sebastian Florentin Hofstätter
Jiecao Chen
Hamed Zamani
arXiv preprint arXiv:2209.14290(2022)
Preview abstract
Retrieval-augmented generation models offer many benefits over standalone language models: besides a textual answer to a given query they provide provenance items retrieved from an updateable knowledge base. However, they are also more complex systems and need to handle long inputs. In this work, we introduce FiD-Light to strongly increase the efficiency of the state-of-the-art retrieval-augmented FiD model, while maintaining the same level of effectiveness. Our FiD-Light model constrains the information flow from the encoder (which encodes passages separately) to the decoder (using concatenated encoded representations). Furthermore, we adapt FiD-Light with re-ranking capabilities through textual source pointers, to improve the top-ranked provenance precision. Our experiments on a diverse set of seven knowledge intensive tasks (KILT) show FiD-Light consistently improves the Pareto frontier between query latency and effectiveness. FiD-Light with source pointing sets substantial new state-of-the-art results on six KILT tasks for combined text generation and provenance retrieval evaluation, while maintaining reasonable efficiency.
View details
Transforming Sequence Tagging Into A Seq2Seq Task
Iftekhar Naim
Jiecao Chen
Kiran Yalasangi
EMNLP 2022
Preview abstract
Pretrained, large, generative language models (LMs) have had great success in a wide range of sequence tagging and structured prediction tasks. Casting a sequence tagging task as a Seq2Seq problem requires deciding the formats of the input and output sequences. However, we lack a principled understanding of the trade-offs associated with these formats (such as the effect on model accuracy, sequence length, multilingual generalization, hallucination). In this paper, we rigorously study different *formats* one could use for casting input text sentences and their output labels into the "input" and "target" of a Seq2Seq model. Along the way, we introduce a new format, which we show to not only be simpler but also more effective. Additionally the new formats demonstrate significant gains in the multilingual settings -- both zero-shot transfer learning and joint training. Lastly, we find that the new formats are more robust and almost completely devoid of the danger of *hallucination* that often plagues existing formats. With well over a 1000 experiments studying 14 different formats, over 7 diverse public benchmarks -- including 3 multilingual datasets spanning 7 languages -- we believe our findings provide a strong empirical basis in understanding how we should tackle sequence tagging tasks.
View details
Preview abstract
Sequential labeling is a fundamental NLP task, forming the backbone of many applications. Supervised learning of Seq2Seq models (like T5) has shown great success on these problems. However there remains a significant disconnect between the training objectives of these models vs the metrics and desiderata we care about in practical applications. For example, a practical sequence tagging application may want to optimize for a certain precision-recall trade-off (of the top-k predictions) which is quite different from the standard objective of maximizing the likelihood of the gold labeled sequence. Thus to bridge this gap, we propose GROOT -- a simple yet effective framework for Generative Reward Optimization Of Text sequences. GROOT works by training a generative sequential labeling model to match the decoder output distribution with that of the (black-box) reward function. Using an iterative training regime, we first generate prediction candidates, then correct errors in them, and finally contrast those candidates (based on their reward values). As demonstrated via extensive experiments on four public benchmarks, GROOT significantly improves all reward metrics. Furthermore, GROOT also leads to improvements of the overall decoder distribution as evidenced by the quality gains of the top- candidates.
View details
Preview abstract
Large Language Models (LLMs) have shown impressive results on a variety of text understanding tasks. Search queries though pose a unique challenge, given their short-length and lack of nuance or context. Complicated feature engineering efforts do not always lead to downstream improvements as their performance benefits may be offset by increased complexity of knowledge distillation. Thus, in this paper we make the following contributions: (1) We demonstrate that Retrieval Augmentation of queries provides LLMs with valuable additional context enabling improved understanding. While Retrieval Augmentation typically increases latency of LMs (thus hurting distillation efficacy), (2) we provide a practical and effective way of distilling Retrieval Augmentation LLMs. Specifically, we use a novel two-stage distillation approach that allows us to carry over the gains of retrieval augmentation, without suffering the increased compute typically associated with it. (3) We demonstrate the benefits of the proposed approach on a billion-scale, real-world query understanding system resulting in an X\% improvement. Via extensive experiments, including on public benchmarks, we believe this work offers a recipe for practical use of retrieval-augmented query understanding.
View details
WIT: Wikipedia-based Image Text Dataset for Multimodal Multilingual Machine Learning
Jiecao Chen
Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR '21)(2021)
Preview abstract
The milestone improvements brought about by deep representation learning and pre-training techniques have led to large performance gains across downstream NLP, IR and Vision tasks. Multimodal modeling techniques aim to leverage high-quality visio-linguistic datasets for learning complementary information (across image and text modalities). In this paper, we introduce the Wikipedia-based Image Text (WIT) Dataset to better facilitate multimodal, multilingual learning. WIT is composed of 11 million+ unique images with over 37 million entity rich text descriptions associated with these images in Wikipedia from over 100 languages. Its size enables WIT to be used as a pretraining dataset for multimodal models, as we show when applied to downstream tasks such as image-text retrieval. WIT has four main and unique advantages. First, WIT is the largest multimodal dataset (at the time of writing). Second, it is massively multilingual (first of its kind) with coverage over 100+ languages (each of which has at least 10K examples) and provides cross-lingual texts for many images. Third, it represents a more diverse set of concepts and real world entities relative to what previous datasets cover. Lastly, as we demonstrate empirically, WIT provides a very challenging real-world test set that empirically highlights the need for learning improvements in tasks such as Retrieval and Captioning.
View details
