Deepak Sharma

Gheorghe Comanici, Eric Bieber, Mike Schaekermann et al. · amazon-science, baidu

In this report, we introduce the Gemini 2.X model family: Gemini 2.5 Pro and Gemini 2.5 Flash, as well as our earlier Gemini 2.0 Flash and Flash-Lite models. Gemini 2.5 Pro is our most capable model yet, achieving SoTA performance on frontier coding and reasoning benchmarks. In addition to its incredible coding and reasoning skills, Gemini 2.5 Pro is a thinking model that excels at multimodal understanding and it is now able to process up to 3 hours of video content. Its unique combination of long context, multimodal and reasoning capabilities can be combined to unlock new agentic workflows. Gemini 2.5 Flash provides excellent reasoning abilities at a fraction of the compute and latency requirements and Gemini 2.0 Flash and Flash-Lite provide high performance at low latency and cost. Taken together, the Gemini 2.X model generation spans the full Pareto frontier of model capability vs cost, allowing users to explore the boundaries of what is possible with complex agentic problem solving.

Larry Bowden, Qi Chu, Bernard Cena et al.

Detecting machine failures promptly is of utmost importance in industry for maintaining efficiency and minimizing downtime. This paper introduces a failure detection algorithm based on quantum computing and a statistical change-point detection approach. Our method leverages the potential of projected quantum feature maps to enhance the precision of anomaly detection in machine monitoring systems. We empirically validate our approach on benchmark multi-dimensional time series datasets as well as on a real-world dataset comprising IoT sensor readings from operational machines, ensuring the practical relevance of our study. The algorithm was executed on IBM's 133-qubit Heron quantum processor, demonstrating the feasibility of integrating quantum computing into industrial maintenance procedures. The presented results underscore the effectiveness of our quantum-based failure detection system, showcasing its capability to accurately identify anomalies in noisy time series data. This work not only highlights the potential of quantum computing in industrial diagnostics but also paves the way for more sophisticated quantum algorithms in the realm of predictive maintenance.

Grishma Sharma, Aditi Paretkar, Deepak Sharma

The availability of a vast array of research papers in any area of study, necessitates the need of automated summarisation systems that can present the key research conducted and their corresponding findings. Scientific paper summarisation is a challenging task for various reasons including token length limits in modern transformer models and corresponding memory and compute requirements for long text. A significant amount of work has been conducted in this area, with approaches that modify the attention mechanisms of existing transformer models and others that utilise discourse information to capture long range dependencies in research papers. In this paper, we propose a hybrid methodology for research paper summarisation which incorporates an extractive and abstractive approach. We use the extractive approach to capture the key findings of research, and pair it with the introduction of the paper which captures the motivation for research. We use two models based on unsupervised learning for the extraction stage and two transformer language models, resulting in four combinations for our hybrid approach. The performances of the models are evaluated on three metrics and we present our findings in this paper. We find that using certain combinations of hyper parameters, it is possible for automated summarisation systems to exceed the abstractiveness of summaries written by humans. Finally, we state our future scope of research in extending this methodology to summarisation of generalised long documents.

Vishal Khare, Vijay Saini, Deepak Sharma et al.

This study investigates the implementation and efficacy of DeputyDev, an AI-powered code review assistant developed to address inefficiencies in the software development process. The process of code review is highly inefficient for several reasons, such as it being a time-consuming process, inconsistent feedback, and review quality not being at par most of the time. Using our telemetry data, we observed that at TATA 1mg, pull request (PR) processing exhibits significant inefficiencies, with average pick-up and review times of 73 and 82 hours, respectively, resulting in a 6.2 day closure cycle. The review cycle was marked by prolonged iterative communication between the reviewing and submitting parties. Research from the University of California, Irvine indicates that interruptions can lead to an average of 23 minutes of lost focus, critically affecting code quality and timely delivery. To address these challenges, we developed DeputyDev's PR review capabilities by providing automated, contextual code reviews. We conducted a rigorous double-controlled A/B experiment involving over 200 engineers to evaluate DeputyDev's impact on review times. The results demonstrated a statistically significant reduction in both average per PR (23.09%) and average per-line-of-code (40.13%) review durations. After implementing safeguards to exclude outliers, DeputyDev has been effectively rolled out across the entire organisation. Additionally, it has been made available to external companies as a Software-as-a-Service (SaaS) solution, currently supporting the daily work of numerous engineering professionals. This study explores the implementation and effectiveness of AI-assisted code reviews in improving development workflow timelines and code.

Talha Chafekar, Aafiya Hussain, Grishma Sharma et al.

There has been a lot of work in question generation where different methods to provide target answers as input, have been employed. This experimentation has been mostly carried out for RNN based models. We use three different methods and their combinations for incorporating answer information and explore their effect on several automatic evaluation metrics. The methods that are used are answer prompting, using a custom product method using answer embeddings and encoder outputs, choosing sentences from the input paragraph that have answer related information, and using a separate cross-attention attention block in the decoder which attends to the answer. We observe that answer prompting without any additional modes obtains the best scores across rouge, meteor scores. Additionally, we use a custom metric to calculate how many of the generated questions have the same answer, as the answer which is used to generate them.

Anand Kumar, Vishal Khare, Deepak Sharma et al.

We present a comprehensive real-world evaluation of AI-assisted software development tools deployed at enterprise scale. Over one year, 300 engineers across multiple teams integrated an in-house AI platform (DeputyDev) that combines code generation and automated review capabilities into their daily workflows. Through rigorous cohort analysis, our study demonstrates statistically significant productivity improvements, including an overall 31.8% reduction in PR review cycle time. Developer adoption was strong, with 85% satisfaction for code review features and 93% expressing a desire to continue using the platform. Adoption patterns showed systematic scaling from 4% engagement in month 1 to 83% peak usage by month 6, stabilizing at 60% active engagement. Top adopters achieved a 61% increase in code volume pushed to production, contributing to approximately 30 to 40% of code shipped to production through this tool, accounting for an overall 28% increase in code shipment volume. Unlike controlled benchmark evaluations, our longitudinal analysis provides empirical evidence from production environments, revealing both the transformative potential and practical deployment challenges of integrating AI into enterprise software development workflows.

Shreya Srivastava, Durgesh Kumar, Ram Jiwari et al.

The variability in ECG readings influenced by individual patient characteristics has posed a considerable challenge to adopting automated ECG analysis in clinical settings. A novel feature fusion technique termed SACC (Self Attentive Canonical Correlation) was proposed to address this. This technique is combined with DPN (Dual Pathway Network) and depth-wise separable convolution to create a robust, interpretable, and fast end-to-end arrhythmia classification model named rECGnition_v2.0 (robust ECG abnormality detection). This study uses MIT-BIH, INCARTDB and EDB dataset to evaluate the efficiency of rECGnition_v2.0 for various classes of arrhythmias. To investigate the influence of constituting model components, various ablation studies were performed, i.e. simple concatenation, CCA and proposed SACC were compared, while the importance of global and local ECG features were tested using DPN rECGnition_v2.0 model and vice versa. It was also benchmarked with state-of-the-art CNN models for overall accuracy vs model parameters, FLOPs, memory requirements, and prediction time. Furthermore, the inner working of the model was interpreted by comparing the activation locations in ECG before and after the SACC layer. rECGnition_v2.0 showed a remarkable accuracy of 98.07% and an F1-score of 98.05% for classifying ten distinct classes of arrhythmia with just 82.7M FLOPs per sample, thereby going beyond the performance metrics of current state-of-the-art (SOTA) models by utilizing MIT-BIH Arrhythmia dataset. Similarly, on INCARTDB and EDB datasets, excellent F1-scores of 98.01% and 96.21% respectively was achieved for AAMI classification. The compact architectural footprint of the rECGnition_v2.0, characterized by its lesser trainable parameters and diminished computational demands, unfurled several advantages including interpretability and scalability.

Gemini Team, Rohan Anil, Sebastian Borgeaud et al.

This report introduces a new family of multimodal models, Gemini, that exhibit remarkable capabilities across image, audio, video, and text understanding. The Gemini family consists of Ultra, Pro, and Nano sizes, suitable for applications ranging from complex reasoning tasks to on-device memory-constrained use-cases. Evaluation on a broad range of benchmarks shows that our most-capable Gemini Ultra model advances the state of the art in 30 of 32 of these benchmarks - notably being the first model to achieve human-expert performance on the well-studied exam benchmark MMLU, and improving the state of the art in every one of the 20 multimodal benchmarks we examined. We believe that the new capabilities of the Gemini family in cross-modal reasoning and language understanding will enable a wide variety of use cases. We discuss our approach toward post-training and deploying Gemini models responsibly to users through services including Gemini, Gemini Advanced, Google AI Studio, and Cloud Vertex AI.

Paul Bertin, Jarrid Rector-Brooks, Deepak Sharma et al.

For large libraries of small molecules, exhaustive combinatorial chemical screens become infeasible to perform when considering a range of disease models, assay conditions, and dose ranges. Deep learning models have achieved state of the art results in silico for the prediction of synergy scores. However, databases of drug combinations are biased towards synergistic agents and these results do not necessarily generalise out of distribution. We employ a sequential model optimization search utilising a deep learning model to quickly discover synergistic drug combinations active against a cancer cell line, requiring substantially less screening than an exhaustive evaluation. Our small scale wet lab experiments only account for evaluation of ~5% of the total search space. After only 3 rounds of ML-guided in vitro experimentation (including a calibration round), we find that the set of drug pairs queried is enriched for highly synergistic combinations; two additional rounds of ML-guided experiments were performed to ensure reproducibility of trends. Remarkably, we rediscover drug combinations later confirmed to be under study within clinical trials. Moreover, we find that drug embeddings generated using only structural information begin to reflect mechanisms of action. Prior in silico benchmarking suggests we can enrich search queries by a factor of ~5-10x for highly synergistic drug combinations by using sequential rounds of evaluation when compared to random selection, or by a factor of >3x when using a pretrained model selecting all drug combinations at a single time point.

Deepak Sharma, Bijendra Kumar, Satish Chand

The aim of this paper is to uncover the researchers in machine learning using the author-topic model (ATM). We collect 16,855 scientific papers from six top journals in the field of machine learning published from 1997 to 2016 and analyze them using ATM. The dataset is broken down into 4 intervals to identify the top researchers and find similar researchers using their similarity score. The similarity score is calculated using Hellinger distance. The researchers are plotted using t-SNE, which reduces the dimensionality of the data while keeping the same distance between the points. The analysis of our study helps the upcoming researchers to find the top researchers in their area of interest.

Deepak Sharma, Audrey Durand, Marc-André Legault et al.

Genome-Wide Association Studies are typically conducted using linear models to find genetic variants associated with common diseases. In these studies, association testing is done on a variant-by-variant basis, possibly missing out on non-linear interaction effects between variants. Deep networks can be used to model these interactions, but they are difficult to train and interpret on large genetic datasets. We propose a method that uses the gradient based deep interpretability technique named DeepLIFT to show that known diabetes genetic risk factors can be identified using deep models along with possibly novel associations.

Siddhant Arora, Srikanta Bedathur, Maya Ramanath et al.

Knowledge Graphs (KGs) extracted from text sources are often noisy and lead to poor performance in downstream application tasks such as KG-based question answering.While much of the recent activity is focused on addressing the sparsity of KGs by using embeddings for inferring new facts, the issue of cleaning up of noise in KGs through KG refinement task is not as actively studied. Most successful techniques for KG refinement make use of inference rules and reasoning over ontologies. Barring a few exceptions, embeddings do not make use of ontological information, and their performance in KG refinement task is not well understood. In this paper, we present a KG refinement framework called IterefinE which iteratively combines the two techniques - one which uses ontological information and inferences rules, PSL-KGI, and the KG embeddings such as ComplEx and ConvE which do not. As a result, IterefinE is able to exploit not only the ontological information to improve the quality of predictions, but also the power of KG embeddings which (implicitly) perform longer chains of reasoning. The IterefinE framework, operates in a co-training mode and results in explicit type-supervised embedding of the refined KG from PSL-KGI which we call as TypeE-X. Our experiments over a range of KG benchmarks show that the embeddings that we produce are able to reject noisy facts from KG and at the same time infer higher quality new facts resulting in up to 9% improvement of overall weighted F1 score

Riashat Islam, Komal K. Teru, Deepak Sharma et al.

Off-policy deep reinforcement learning (RL) algorithms are incapable of learning solely from batch offline data without online interactions with the environment, due to the phenomenon known as \textit{extrapolation error}. This is often due to past data available in the replay buffer that may be quite different from the data distribution under the current policy. We argue that most off-policy learning methods fundamentally suffer from a \textit{state distribution shift} due to the mismatch between the state visitation distribution of the data collected by the behavior and target policies. This data distribution shift between current and past samples can significantly impact the performance of most modern off-policy based policy optimization algorithms. In this work, we first do a systematic analysis of state distribution mismatch in off-policy learning, and then develop a novel off-policy policy optimization method to constraint the state distribution shift. To do this, we first estimate the state distribution based on features of the state, using a density estimator and then develop a novel constrained off-policy gradient objective that minimizes the state distribution shift. Our experimental results on continuous control tasks show that minimizing this distribution mismatch can significantly improve performance in most popular practical off-policy policy gradient algorithms.

Shefali Srivastava, Abhimanyu Chopra, Arun CS Kumar et al.

An end-to-end trainable ConvNet architecture, that learns to harness the power of shape representation for matching disparate image pairs, is proposed. Disparate image pairs are deemed those that exhibit strong affine variations in scale, viewpoint and projection parameters accompanied by the presence of partial or complete occlusion of objects and extreme variations in ambient illumination. Under these challenging conditions, neither local nor global feature-based image matching methods, when used in isolation, have been observed to be effective. The proposed correspondence determination scheme for matching disparate images exploits high-level shape cues that are derived from low-level local feature descriptors, thus combining the best of both worlds. A graph-based representation for the disparate image pair is generated by constructing an affinity matrix that embeds the distances between feature points in two images, thus modeling the correspondence determination problem as one of graph matching. The eigenspectrum of the affinity matrix, i.e., the learned global shape representation, is then used to further regress the transformation or homography that defines the correspondence between the source image and target image. The proposed scheme is shown to yield state-of-the-art results for both, coarse-level shape matching as well as fine point-wise correspondence determination.

Sachin Lakra, T. V. Prasad, Deepak Sharma et al.

A Neural Network, in general, is not considered to be a good solver of mathematical and binary arithmetic problems. However, networks have been developed for such problems as the XOR circuit. This paper presents a technique for the implementation of the Half-adder circuit using the CoActive Neuro-Fuzzy Inference System (CANFIS) Model and attempts to solve the problem using the NeuroSolutions 5 Simulator. The paper gives the experimental results along with the interpretations and possible applications of the technique.