arXiv subjects

Nilesh Jain

Publications and source records attributed to Nilesh Jain.

At least 19 recordsLinked to original sources

arXiv2026

FOLIO: Focused Semantic Memory for Streaming Video Understanding

In online streaming video understanding, a video stream continues to arrive and queries may be issued at any time. Because streaming frames grow without bound, the system must continuously compress and retain information from the observed video prefix while future frames and future queries remain unknown. The core challenge is deciding what information to retain and how to organize the maintained history: as this history grows with the stream, memory cost increases and many redundant visual details are retained, whereas later queries often depend on specific entities, actions, and their temporal changes. To address this challenge, we introduce FOLIO, a training-free focused semantic memory system that records important parts of the stream in higher detail while keeping surrounding context compact. As the stream arrives, FOLIO updates memory at the segment level, guided by a dynamic focus state, combining a short-term visual buffer with a long-term semantic memory organized around observed entities and linked to a visual-evidence cache. At query time, lightweight hybrid retrieval combines direct matching over the structured memory with semantic query expansion. FOLIO achieves state-of-the-art performance, reaching 82.0/69.1 Perception/Backward accuracy on OVO-Bench with Qwen3-VL-8B and 74.5 overall accuracy on StreamingBench, while substantially reducing the cost of maintaining streaming memory by reserving detailed records for focused entities and storing surrounding context compactly.

cs.CV↗

arXiv2026

Bibby AI: An Editor-Native Agentic Platform for Academic Research, Writing, and Publishing

Academic output is produced across a fragmented toolchain: literature discovery in one application, reference management in another, writing in a LaTeX editor, formatting against venue templates by hand, and submission through yet another portal. Each boundary between tools forces a context switch, a format conversion, or a manual copy-paste step, and the cumulative cost dominates the time researchers spend on activities that are not research. We present Bibby AI, an editor-native platform that collapses this toolchain into a single Research-Write-Publish pipeline built around a cloud LaTeX editor. Unlike assistants that attach to an existing editor through a browser extension, Bibby AI owns the full document state, compilation pipeline, and revision history, which allows its agents to perform retrieval-grounded citation insertion, structural edits, and template-compliant reformatting as first-class, verifiable operations rather than text suggestions. The platform integrates (i) ingestion pipelines that convert PDF, DOCX, and handwritten mathematics into clean LaTeX; (ii) a retrieval layer over scholarly metadata enriched with patent-to-paper citation signals derived from USPTO PatentsView and the Marx-Fuegi citation corpus, surfacing the translational impact of candidate references; and (iii) task-scoped agents for literature triage, drafting, revision, and venue formatting that operate directly on the document's abstract syntax representation. Bibby AI is deployed in production and serves more than 5,000 active researchers across more than 50 subscribing universities. We describe the architecture, the design decisions that editor-nativeness makes possible, and the workflow-level time-savings framework we use to evaluate the platform against fragmented baselines.

cs.DL↗

arXiv2026

LLM-Guided ANN Index Optimization for Human-Object Interaction Retrieval

Retrieval systems underpin modern AI applications -- spanning visual search, recommendation engines, and multi-modal question answering. Modern multi-stage retrieval systems require the joint optimization of highly coupled parameters, yet traditional hyperparameter optimization (HPO) methods -- including Tree-structured Parzen Estimators (TPE) and Gaussian Process Bayesian Optimization -- rely on an independence assumption that fundamentally prevents them from navigating these coupled configuration spaces. We address this limitation with a phase-aware large language model (LLM) agent that conditions each proposal on its full optimization history, navigating the coupled parameter space across phase-partitioned exploration, exploitation, and fine-tuning stages. Evaluated on the HICO-DET human-object interaction retrieval benchmark using Intel VDMS (Visual Data Management System), our agent outperforms Optuna TPE by +33.3% and VDTuner by +34.2% under SIEVE (Safeguarded Index Evaluation of Vector-search Efficiency, a quality-constrained throughput metric), delivering a 15.3x throughput gain over UniIR. Validation across three benchmarks confirms that the agent's advantage grows with the degree of parameter coupling: +33.3% on HICO-DET (high coupling), methods converge within 1% on GLDv2 (moderate coupling) and within 3.6% on SIFT1M (near-independent control). Cross-system validation on Milvus confirms the optimizer ranks first on all three datasets without modification, demonstrating transferability across vector database management system (VDBMS) platforms.

cs.CV↗

arXiv2026

TokenButler: Token Importance is Predictable

Large Language Models (LLMs) rely on the Key-Value (KV) Cache to store token history, enabling efficient decoding of tokens. As the KV-Cache grows, it becomes a major memory and computation bottleneck. However, there is an opportunity to alleviate this bottleneck, prior research has shown that only a small subset of tokens contribute meaningfully to each decoding step. A key challenge in finding these critical tokens is that they are dynamic, and heavily input query-dependent. Existing methods either risk quality by evicting tokens permanently, or retain the full KV-Cache but rely on retrieving chunks of tokens and many existing KV-Cache sparsity methods rely on inaccurate proxies for token importance. To address these limitations, we introduce TokenButler, a high-granularity, query-aware predictor that learns to identify these critical tokens. TokenButler predicts low-dimensional importance queries at a fixed depth stride, and combines them with a learned projection of the real KV-cache keys to score tokens cheaply, enabling dynamic per-token selection under a fixed budget while preserving the full KV cache. We train TokenButler by distilling the model's masked causal attention distributions, optimizing a lightweight predictor with minimal parameter overhead. We evaluate TokenButler on a novel synthetic small-context co-referential retrieval task, demonstrating near-oracle accuracy where existing methods fail. Furthermore, TokenButler achieves competitive or superior performance on long-context benchmarks (RULER, LongBench), up to $\approx1.6\times$ on-GPU speedup using our proposed *prediction interval with neighbor fetching* that amortizes predictor cost while maintaining accuracy within $\approx$1.1\%, and up to 7.6$\times$ reduction in latency compared to Dense Attention with CPU offloading. Code is available: https://github.com/abdelfattah-lab/TokenButler

cs.CL↗

arXiv2026

AutoResearch-RL: Perpetual Self-Evaluating Reinforcement Learning Agents for Autonomous Neural Architecture Discovery

We present AutoResearch-RL, a framework in which a reinforcement learning agent conducts open-ended neural architecture and hyperparameter research without human supervision, running perpetually until a termination oracle signals convergence or resource exhaustion. At each step the agent proposes a code modification to a target training script, executes it under a fixed wall clock time budget, observes a scalar reward derived from validation bits-per-byte (val-bpb), and updates its policy via Proximal Policy Optimisation (PPO). The key design insight is the separation of three concerns: (i) a frozen environment (data pipeline, evaluation protocol, and constants) that guarantees fair cross-experiment comparison; (ii) a mutable target file (train.py) that represents the agent's editable state; and (iii) a meta-learner (the RL agent itself) that accumulates a growing trajectory of experiment outcomes and uses them to inform subsequent proposals. We formalise this as a Markov Decision Process, derive convergence guarantees under mild assumptions, and demonstrate empirically on a single GPU nanochat pretraining benchmark that AutoResearch-RL discovers configurations that match or exceed hand-tuned baselines after approximately 300 overnight iterations, with no human in the loop.

cs.LG↗

arXiv2026

Multi-LLM Thematic Analysis with Dual Reliability Metrics: Combining Cohen's Kappa and Semantic Similarity for Qualitative Research Validation

Qualitative research faces a critical reliability challenge: traditional inter-rater agreement methods require multiple human coders, are time-intensive, and often yield moderate consistency. We present a multi-perspective validation framework for LLM-based thematic analysis that combines ensemble validation with dual reliability metrics: Cohen's Kappa ($κ$) for inter-rater agreement and cosine similarity for semantic consistency. Our framework enables configurable analysis parameters (1-6 seeds, temperature 0.0-2.0), supports custom prompt structures with variable substitution, and provides consensus theme extraction across any JSON format. As proof-of-concept, we evaluate three leading LLMs (Gemini 2.5 Pro, GPT-4o, Claude 3.5 Sonnet) on a psychedelic art therapy interview transcript, conducting six independent runs per model. Results demonstrate Gemini achieves highest reliability ($κ= 0.907$, cosine=95.3%), followed by GPT-4o ($κ= 0.853$, cosine=92.6%) and Claude ($κ= 0.842$, cosine=92.1%). All three models achieve a high agreement ($κ> 0.80$), validating the multi-run ensemble approach. The framework successfully extracts consensus themes across runs, with Gemini identifying 6 consensus themes (50-83% consistency), GPT-4o identifying 5 themes, and Claude 4 themes. Our open-source implementation provides researchers with transparent reliability metrics, flexible configuration, and structure-agnostic consensus extraction, establishing methodological foundations for reliable AI-assisted qualitative research.

cs.CL↗

arXiv2026

Mem-Rec: Memory Efficient Recommendation System using Alternative Representation

Deep learning-based recommendation systems (e.g., DLRMs) are widely used AI models to provide high-quality personalized recommendations. Training data used for modern recommendation systems commonly includes categorical features taking on tens-of-millions of possible distinct values. These categorical tokens are typically assigned learned vector representations, that are stored in large embedding tables, on the order of 100s of GB. Storing and accessing these tables represent a substantial burden in commercial deployments. Our work proposes MEM-REC, a novel alternative representation approach for embedding tables. MEM-REC leverages bloom filters and hashing methods to encode categorical features using two cache-friendly embedding tables. The first table (token embedding) contains raw embeddings (i.e. learned vector representation), and the second table (weight embedding), which is much smaller, contains weights to scale these raw embeddings to provide better discriminative capability to each data point. We provide a detailed architecture, design and analysis of MEM-REC addressing trade-offs in accuracy and computation requirements, in comparison with state-of-the-art techniques. We show that MEM-REC can not only maintain the recommendation quality and significantly reduce the memory footprint for commercial scale recommendation models but can also improve the embedding latency. In particular, based on our results, MEM-REC compresses the MLPerf CriteoTB benchmark DLRM model size by 2900x and performs up to 3.4x faster embeddings while achieving the same AUC as that of the full uncompressed model.

cs.IR↗

arXiv2026

KVCrush: Key value cache size-reduction using similarity in head-behaviour

Key-value (KV) caching has emerged as a crucial optimization technique for accelerating inference in large language models (LLMs). By allowing the attention operation to scale linearly rather than quadratically with the total sequence length, KV caching significantly enhances generation throughput. However, due to large context lengths in the modern LLMs, the memory footprint of the KV is a huge bottleneck for model deployment directly impacting the model's batch size, hindering its ability to deliver high-throughput. Existing research addresses this challenge using several techniques, such as discarding low-attention tokens, quantization, and matrix approximation which typically lead to a negative impact on the model accuracy. In this paper, We propose KVCrush technology which can be combined with many KV compression technologies to improve the model accuracy at a much smaller memory. KVCrush provides an alternate representation scheme for key-value states, along with a low-overhead token pruning algorithm that accounts for the token distribution in the KV cache, which in turn allows for a a smaller footprint while maintaining the accuracy of the model. Based on our results, KVCrush reduces LongBench KV Cache size by 4x with less than 1% accuracy drop and achieves state-of-the-art average accuracy with minimal overhead, incurring less than 0.5% total inference latency. KVCrush not only outperforms the accuracy of state-of-the-art importance-based token retention schemes but is also compatible with typical practical LLM deployments using KV cache paging schemes such as vLLM and mixed precision quantization.

cs.CL↗

arXiv2025

Denoising Diffusion as a New Framework for Underwater Images

Underwater images play a crucial role in ocean research and marine environmental monitoring since they provide quality information about the ecosystem. However, the complex and remote nature of the environment results in poor image quality with issues such as low visibility, blurry textures, color distortion, and noise. In recent years, research in image enhancement has proven to be effective but also presents its own limitations, like poor generalization and heavy reliance on clean datasets. One of the challenges herein is the lack of diversity and the low quality of images included in these datasets. Also, most existing datasets consist only of monocular images, a fact that limits the representation of different lighting conditions and angles. In this paper, we propose a new plan of action to overcome these limitations. On one hand, we call for expanding the datasets using a denoising diffusion model to include a variety of image types such as stereo, wide-angled, macro, and close-up images. On the other hand, we recommend enhancing the images using Controlnet to evaluate and increase the quality of the corresponding datasets, and hence improve the study of the marine ecosystem. Tags - Underwater Images, Denoising Diffusion, Marine ecosystem, Controlnet

cs.CV↗

arXiv2025

ContraGS: Codebook-Condensed and Trainable Gaussian Splatting for Fast, Memory-Efficient Reconstruction

3D Gaussian Splatting (3DGS) is a state-of-art technique to model real-world scenes with high quality and real-time rendering. Typically, a higher quality representation can be achieved by using a large number of 3D Gaussians. However, using large 3D Gaussian counts significantly increases the GPU device memory for storing model parameters. A large model thus requires powerful GPUs with high memory capacities for training and has slower training/rendering latencies due to the inefficiencies of memory access and data movement. In this work, we introduce ContraGS, a method to enable training directly on compressed 3DGS representations without reducing the Gaussian Counts, and thus with a little loss in model quality. ContraGS leverages codebooks to compactly store a set of Gaussian parameter vectors throughout the training process, thereby significantly reducing memory consumption. While codebooks have been demonstrated to be highly effective at compressing fully trained 3DGS models, directly training using codebook representations is an unsolved challenge. ContraGS solves the problem of learning non-differentiable parameters in codebook-compressed representations by posing parameter estimation as a Bayesian inference problem. To this end, ContraGS provides a framework that effectively uses MCMC sampling to sample over a posterior distribution of these compressed representations. With ContraGS, we demonstrate that ContraGS significantly reduces the peak memory during training (on average 3.49X) and accelerated training and rendering (1.36X and 1.88X on average, respectively), while retraining close to state-of-art quality.

cs.GR↗

arXiv2025

SparAMX: Accelerating Compressed LLMs Token Generation on AMX-powered CPUs

Large language models have high compute, latency, and memory requirements. While specialized accelerators such as GPUs and TPUs typically run these workloads, CPUs are more widely available and consume less energy. Accelerating LLMs with CPUs enables broader AI access at a lower cost and power consumption. This acceleration potential for CPUs is especially relevant during the memory-bound decoding stage of LLM inference, which processes one token at a time and is becoming increasingly utilized with reasoning models. We utilize Advanced Matrix Extensions (AMX) support on the latest Intel CPUs together with unstructured sparsity to achieve a $1.42 \times$ reduction in end-to-end latency compared to the current PyTorch implementation by applying our technique in linear layers. We provide a set of open-source customized sparse kernels that can speed up any PyTorch model by automatically replacing all linear layers with our custom sparse implementation. Furthermore, we demonstrate for the first time the use of unstructured sparsity in the attention computation achieving a $1.14 \times$ speedup over the current systems without compromising accuracy. Code: https://github.com/IntelLabs/Hardware-Aware-Automated-Machine-Learning/tree/main/SparAMX

cs.LG↗

arXiv2025

Mamba-Shedder: Post-Transformer Compression for Efficient Selective Structured State Space Models

Large pre-trained models have achieved outstanding results in sequence modeling. The Transformer block and its attention mechanism have been the main drivers of the success of these models. Recently, alternative architectures, such as Selective Structured State Space Models (SSMs), have been proposed to address the inefficiencies of Transformers. This paper explores the compression of SSM-based models, particularly Mamba and its hybrids. We study the sensitivity of these models to the removal of selected components at different granularities to reduce the model size and computational overhead, thus improving their efficiency while maintaining accuracy. The proposed solutions, collectively referred to as Mamba-Shedder, achieve a speedup of up to 1.4x during inference, demonstrating that model efficiency can be improved by eliminating several redundancies with minimal impact on the overall model performance. The code is available at https://github.com/IntelLabs/Hardware-Aware-Automated-Machine-Learning.

cs.LG↗

arXiv2025

INRet: A General Framework for Accurate Retrieval of INRs for Shapes

Implicit neural representations (INRs) have become an important method for encoding various data types, such as 3D objects or scenes, images, and videos. They have proven to be particularly effective at representing 3D content, e.g., 3D scene reconstruction from 2D images, novel 3D content creation, as well as the representation, interpolation, and completion of 3D shapes. With the widespread generation of 3D data in an INR format, there is a need to support effective organization and retrieval of INRs saved in a data store. A key aspect of retrieval and clustering of INRs in a data store is the formulation of similarity between INRs that would, for example, enable retrieval of similar INRs using a query INR. In this work, we propose INRet, a method for determining similarity between INRs that represent shapes, thus enabling accurate retrieval of similar shape INRs from an INR data store. INRet flexibly supports different INR architectures such as INRs with octree grids, triplanes, and hash grids, as well as different implicit functions including signed/unsigned distance function and occupancy field. We demonstrate that our method is more general and accurate than the existing INR retrieval method, which only supports simple MLP INRs and requires the same architecture between the query and stored INRs. Furthermore, compared to converting INRs to other representations (e.g., point clouds or multi-view images) for 3D shape retrieval, INRet achieves higher accuracy while avoiding the conversion overhead.

cs.LG↗

arXiv2025

Low-Rank Adapters Meet Neural Architecture Search for LLM Compression

The rapid expansion of Large Language Models (LLMs) has posed significant challenges regarding the computational resources required for fine-tuning and deployment. Recent advancements in low-rank adapters have demonstrated their efficacy in parameter-efficient fine-tuning (PEFT) of these models. This retrospective paper comprehensively discusses innovative approaches that synergize low-rank representations with Neural Architecture Search (NAS) techniques, particularly weight-sharing super-networks. Robust solutions for compressing and fine-tuning large pre-trained models are developed by integrating these methodologies. Our analysis highlights the potential of these combined strategies to democratize the use of LLMs, making them more accessible for deployment in resource-constrained environments. The resulting models exhibit reduced memory footprints and faster inference times, paving the way for more practical and scalable applications of LLMs. Models and code are available at https://github.com/IntelLabs/Hardware-Aware-Automated-Machine-Learning.

cs.LG↗

arXiv2025

MultiPruner: Balanced Structure Removal in Foundation Models

Recently, state-of-the-art approaches for pruning large pre-trained models (LPMs) have demonstrated that the training-free removal of non-critical residual blocks in Transformers is viable for reducing model size, achieving results that outperform previous training-free pruning approaches. Motivated by these findings, we extend BlockPruner (Zhong et al., 2024) and propose MultiPruner, a pruning approach that surpasses recent training-free pruning methods by adopting a multidimensional, iterative, fine-grained pruning strategy. In MultiPruner, multidimensional pruning reinstates the structural balance in block-pruned models by sequentially compressing along three dimensions: i) residual blocks, ii) channels of multilayer perceptrons (MLP), and iii) attention heads. This solution enhances zero-shot accuracy on downstream tasks compared to other techniques while improving model compression ratios, producing compressed models with fewer computing and memory requirements. Extensive experiments demonstrate the advantages of the proposed method across various large pre-trained models. The code and pruning configurations are available at https://github.com/IntelLabs/Hardware-Aware-Automated-Machine-Learning.

cs.LG↗

arXiv2025

Enhancing Data Integrity through Provenance Tracking in Semantic Web Frameworks

This paper explores the integration of provenance tracking systems within the context of Semantic Web technologies to enhance data integrity in diverse operational environments. SURROUND Australia Pty Ltd demonstrates innovative applica-tions of the PROV Data Model (PROV-DM) and its Semantic Web variant, PROV-O, to systematically record and manage provenance information across multiple data processing domains. By employing RDF and Knowledge Graphs, SURROUND ad-dresses the critical challenges of shared entity identification and provenance granularity. The paper highlights the company's architecture for capturing comprehensive provenance data, en-abling robust validation, traceability, and knowledge inference. Through the examination of two projects, we illustrate how provenance mechanisms not only improve data reliability but also facilitate seamless integration across heterogeneous systems. Our findings underscore the importance of sophisticated provenance solutions in maintaining data integrity, serving as a reference for industry peers and academics engaged in provenance research and implementation.

cs.CR↗

arXiv2024

Post-Training Statistical Calibration for Higher Activation Sparsity

We present Statistical Calibrated Activation Pruning (SCAP), a post-training activation pruning framework that (1) generalizes sparsification by input activations of Fully-Connected layers for generic and flexible application across Transformers, and (2) features a simple Mode-Centering technique to pre-calibrate activation distributions for maximizing post-training sparsity. Our results demonstrate robust Pareto efficiency compared to prior methods, translating to a 1.5x additional LLM decoding speedup against CATS at iso model quality. SCAP effectiveness is empirically verified across a wide range of models, including recent Transformer Decoders, MoE, Mamba2, Encoding Transformer, and pre-quantized models, highlighting its practicality and scalability. The code is available at: https://github.com/IntelLabs/SCAP.

cs.LG↗

arXiv2024

SQFT: Low-cost Model Adaptation in Low-precision Sparse Foundation Models

Large pre-trained models (LPMs), such as large language models, have become ubiquitous and are employed in many applications. These models are often adapted to a desired domain or downstream task through a fine-tuning stage. This paper proposes SQFT, an end-to-end solution for low-precision sparse parameter-efficient fine-tuning of LPMs, allowing for effective model manipulation in resource-constrained environments. Additionally, an innovative strategy enables the merging of sparse weights with low-rank adapters without losing sparsity and accuracy, overcoming the limitations of previous approaches. SQFT also addresses the challenge of having quantized weights and adapters with different numerical precisions, enabling merging in the desired numerical format without sacrificing accuracy. Multiple adaptation scenarios, models, and comprehensive sparsity levels demonstrate the effectiveness of SQFT. Models and code are available at https://github.com/IntelLabs/Hardware-Aware-Automated-Machine-Learning.

cs.LG↗