arXiv subjects

Saurabh Gupta

Publications and source records attributed to Saurabh Gupta.

At least 19 recordsLinked to original sources

arXiv2026

Fine-Tuning VLAs with Self-Demonstrated Generative Control for Multi-Task Manipulation

State-of-the-art vision-language-action (VLA) models such as $\pi_{0.5}$ exhibit strong semantic understanding, instruction following and task behavior. However, when deployed on new robots, even minor mismatches in hardware configuration relative to pretraining can cause severe performance drops. Finetuning the VLA on in-domain expert data from the new embodiment improves performance on the expert task but leads to a loss in its original instruction following and behavioral priors. In this paper, we propose a self-supervised method that generates online interaction rollouts from the zero-shot VLA as additional training data for finetuning. Our experiments show this finetuning scheme yields strong multi-task policies that, on the target robot, (1) inherit prior tasks distilled from the zero-shot model, (2) enable generalist instruction following, while (3) learning new skills from expert data with improved sample efficiency. We demonstrate the success of our approach across test sets probing generalization on a real ALOHA robot and a new simulation benchmark in RoboTwin. Video results are available at https://self-supervised-control.pages.dev/

cs.RO↗

arXiv2026

Improving Map Consistency in Graph-Based LiDAR SLAM Through Information-Aware Odometry and Retroactive Loop Closure

High-quality maps are fundamental for robotics tasks such as navigation and planning. Although modern graph-based LiDAR SLAM systems achieve good trajectory accuracies, a low trajectory error alone does not guarantee geometrically consistent maps, particularly at revisit locations where missed loop closures and residual drift can produce local misalignments. In this work, we address the problem of jointly improving global trajectory estimation and local map quality in 3D LiDAR SLAM. We first propose a framework to efficiently estimate geometry-dependent information matrices for ICP, enabling principled weighting of odometry constraints in a pose graph. We then introduce a hierarchical loop-closure module that decouples place recognition from geometric registration, together with a retroactive loop-closure module that exploits the optimized pose graph to recover missed loop closures. We also propose an evaluation protocol to measure map consistency at revisit locations. We evaluate our SLAM system on several datasets against state-of-the-art LiDAR SLAM systems. Experimental results demonstrate global trajectory accuracies on par with or better than existing methods while consistently improving local geometric map consistency at revisit locations. These results suggest that coupling uncertainty-aware odometry with geometry-guided loop-closure refinement leads to more accurate trajectories and higher-quality maps.

cs.RO↗

arXiv2026

ContactMimic: Humanoid Object Interaction via Contact Control

Keypoint tracking alone is insufficient for object interaction tasks such as sitting on a chair, wiping a board, or pushing furniture, where the robot can reach the correct pose without making meaningful physical contact with the object. We present CONTACTMIMIC, a learning framework that tracks explicit partlevel binary contact commands alongside keypoint trajectories. CONTACTMIMIC is made possible through the use of contact-following rewards and a trajectory augmentation scheme aimed at breaking the correlations between keypoint trajectories and contact labels. The resulting policy successfully decouples contact behavior from keypoint geometry, and achieves precise physical contact as well as contact-controllability (produce or suppress contact during deployment as desired). Simulation experiments across 10 diverse human-object interaction motions confirm that CONTACTMIMIC exhibits contact controllability that enables it to complete manipulation tasks without task-specific rewards, while also outperforming keypoint-only trackers on contact-relevant tasks. Ablations confirm the necessity of the proposed trajectory augmentation scheme and sim2real deployment validates contact controllability in the real world across 5 different motions. Video results are available on https://lixinyao11.github.io/contactmimic-page/.

cs.RO↗

arXiv2026

Unified Semantic Modeling Framework for Large-Scale Job Understanding at LinkedIn

Job understanding is critical to LinkedIn's mission of connecting talent with opportunity. This task involves transforming unstructured and noisy job postings into standardized or derived job attributes that power numerous LinkedIn products. However, building a scalable, cost-efficient, and high-performing job understanding system remains challenging. In this paper, we present a unified semantic modeling framework powered by a small language model (SLM) to address the challenges. We begin by fine-tuning an open-source SLM using a suite of carefully curated synthetic tasks augmented with reasoning traces. These tasks jointly target taxonomy-guided classification and taxonomy-agnostic entity extraction. This allows the resulting model to acquire robust zero-shot generalization for job understanding in structured and unstructured contexts. Building upon this foundation, we introduce a multi-adapter architecture with attribute grouping to facilitate efficient task-specific adaptation while streamlining model management across diverse downstream attributes. Offline evaluations and online A/B tests demonstrate significant performance improvement while reducing operational complexity. Our work provides practical insights into building industry-scale text understanding systems.

cs.AI↗

arXiv2026

Correlation Is Not Enough: Embedding Human Metadata for Individual Causal Discovery

Ask a pretrained biomedical language model whether "cortisol 28 ug/dL" and "stock-market volatility" are related, and it returns a cosine similarity of 0.83 on a scale where 1.0 means identical. The two share no mechanism. This is not a corner case: every off-the-shelf biomedical encoder we tested (BioBERT, PubMedBERT, BioM-ELECTRA) scores unrelated cross-domain pairs between 0.76 and 0.92 when the answer should be near zero. Accuracy on cross-domain discrimination is 0%. Retrieval systems survive this, because a language model downstream filters the noise. A Large Behavioural Model (LBM), a foundation model whose subject is a person rather than a sentence, does not: it reasons over a graph of a user's life and treats embedding proximity as evidence that two events are causally linked. False proximity writes a false causal edge, and everything downstream inherits the error. Here, embedding geometry is not a tuning knob; it is correctness. We report the fix. A contrastive pass over 72,034 pairs raises PubMedBERT BIOSSES correlation from 0.633 to 0.828 and within-vs-across-domain separation from 1.05x to 1.63x. A second pass, BODHI, mines hard negatives from edges absent in a biomedical knowledge graph and lifts separation to 2.30x and the discrimination gap to +0.392, at a 4.5% BIOSSES cost. On an Intel Xeon 6737P with AMX, OpenVINO cuts single-query latency from 1367 ms to 10 ms (133x) and reaches 555 sentences/sec. One finding contradicts standard advice: FP16 beats INT8 on this silicon at every serving batch size, and we explain why. The same model on a no-AMX Ice Lake instance runs 13-27x slower. We release the benchmark suite, training corpora, the BODHI generator, and the OpenVINO scripts.

cs.AI↗

arXiv2026

What Matters When Cotraining Robot Manipulation Policies on Everyday Human Videos?

Human video datasets used for cotraining robot manipulation policies largely consist of curated demonstrations where motions are orchestrated to resemble robot behavior and 3D hand poses are captured with specialized hardware. A more plentiful source of data is everyday Internet video, but it is an open question what factors enable transfer from such videos to robots. We investigate this using a new dataset of 532 human videos with 28 hours of high-quality triangulated hand labels and natural motions. We find that hand pose quality affects transfer, but even with accurate hands, the inherent motion gap hinders transfer unless the vision and policy networks specialize to each embodiment. Our cotraining recipe yields consistent improvements, with an absolute success rate gain of $29.7\%$ in the low-robot-data regime across six manipulation tasks.

cs.RO↗

arXiv2026

MoRight: Motion Control Done Right

Generating motion-controlled videos--where user-specified actions drive physically plausible scene dynamics under freely chosen viewpoints--demands two capabilities: (1) disentangled motion control, allowing users to separately control the object motion and adjust camera viewpoint; and (2) motion causality, ensuring that user-driven actions trigger coherent reactions from other objects rather than merely displacing pixels. Existing methods fall short on both fronts: they entangle camera and object motion into a single tracking signal and treat motion as kinematic displacement without modeling causal relationships between object motion. We introduce MoRight, a unified framework that addresses both limitations through disentangled motion modeling. Object motion is specified in a canonical static-view and transferred to an arbitrary target camera viewpoint via temporal cross-view attention, enabling disentangled camera and object control. We further decompose motion into active (user-driven) and passive (consequence) components, training the model to learn motion causality from data. At inference, users can either supply active motion and MoRight predicts consequences (forward reasoning), or specify desired passive outcomes and MoRight recovers plausible driving actions (inverse reasoning), all while freely adjusting the camera viewpoint. Experiments on three benchmarks demonstrate state-of-the-art performance in generation quality, motion controllability, and interaction awareness.

cs.CV↗

arXiv2026

Loop Closure via Maximal Cliques in 3D LiDAR-Based SLAM

Reliable loop closure detection remains a critical challenge in 3D LiDAR-based SLAM, especially under sensor noise, environmental ambiguity, and viewpoint variation conditions. RANSAC is often used in the context of loop closures for geometric model fitting in the presence of outliers. However, this approach may fail, leading to map inconsistency. We introduce a novel deterministic algorithm, CliReg, for loop closure validation that replaces RANSAC verification with a maximal clique search over a compatibility graph of feature correspondences. This formulation avoids random sampling and increases robustness in the presence of noise and outliers. We integrated our approach into a real- time pipeline employing binary 3D descriptors and a Hamming distance embedding binary search tree-based matching. We evaluated it on multiple real-world datasets featuring diverse LiDAR sensors. The results demonstrate that our proposed technique consistently achieves a lower pose error and more reliable loop closures than RANSAC, especially in sparse or ambiguous conditions. Additional experiments on 2D projection-based maps confirm its generality across spatial domains, making our approach a robust and efficient alternative for loop closure detection.

cs.RO↗

arXiv2026

HERO: Learning Humanoid End-Effector Control for Visual Whole-Body Open-Vocabulary Object Grasping

Visual loco-manipulation of arbitrary in-the-wild objects requires accurate end-effector (EE) control and a generalizable understanding of the scene from visual inputs (eg, RGB-D images). Existing imitation and sim2real methods jointly learn both these aspects via monolithic end-to-end learning and are thus hard to scale. In this work, we bring to bear the best tools for each of these problems -- large vision models for generalizable scene understanding and simulated training for accurate EE control -- leading to an overall modular loco-manipulation system that exhibits strong generalization. Our core technical innovation is HERO, an accurate residual-aware EE tracking policy made possible by combining classical robotics with machine learning. It uses a) inverse kinematics to convert residual end-effector targets into reference trajectories, b) a learned neural forward model for accurate forward kinematics, and c) goal adjustment and replanning. Together, these innovations reduce the end-effector tracking error to 2.44cm, outperforming the strongest prior method by 5.5x. Our overall system operates in diverse real-world environments, from offices to coffee shops, where the robot reliably grasps various everyday objects (eg, mugs, apples, toys) on surfaces ranging from 43cm to 92cm in height. Systematic modular and end-to-end tests demonstrate the effectiveness of our proposed design. We believe our advances open up new ways of training humanoids to interact with daily objects.

cs.RO↗

arXiv2025

Visual Sync: Multi-Camera Synchronization via Cross-View Object Motion

Today, people can easily record memorable moments, ranging from concerts, sports events, lectures, family gatherings, and birthday parties with multiple consumer cameras. However, synchronizing these cross-camera streams remains challenging. Existing methods assume controlled settings, specific targets, manual correction, or costly hardware. We present VisualSync, an optimization framework based on multi-view dynamics that aligns unposed, unsynchronized videos at millisecond accuracy. Our key insight is that any moving 3D point, when co-visible in two cameras, obeys epipolar constraints once properly synchronized. To exploit this, VisualSync leverages off-the-shelf 3D reconstruction, feature matching, and dense tracking to extract tracklets, relative poses, and cross-view correspondences. It then jointly minimizes the epipolar error to estimate each camera's time offset. Experiments on four diverse, challenging datasets show that VisualSync outperforms baseline methods, achieving an median synchronization error below 50 ms.

cs.CV↗

arXiv2025

Quantum theory of electrically levitated nanoparticle-ion systems: Motional dynamics and sympathetic cooling

We develop the theory describing the quantum coupled dynamics of the center-of-mass motion of a nanoparticle and an ensemble of ions co-trapped in a dual-frequency linear Paul trap. We first derive analytical expressions for the motional frequencies and classical trajectories of both nanoparticle and ions. We then derive a quantum master equation for the ion-nanoparticle system and quantify the sympathetic cooling of the nanoparticle motion enabled by its Coulomb coupling to a continuously Doppler-cooled ion. We predict that motional cooling down to sub-kelvin temperatures is achievable in state-of-the-art experiments even in the absence of motional feedback and in the presence of micromotion. We then extend our analysis to an ensemble of $N$ ions, predicting a linear increase of the cooling rate as a function of $N$ and motional cooling of the nanoparticle down to tenths of millikelvin in current experimental platforms. Our work establishes the theoretical toolbox needed to explore the ion-assisted preparation of non-Gaussian motional states of levitated nanoparticles.

quant-ph↗

arXiv2025

Bimanual 3D Hand Motion and Articulation Forecasting in Everyday Images

We tackle the problem of forecasting bimanual 3D hand motion & articulation from a single image in everyday settings. To address the lack of 3D hand annotations in diverse settings, we design an annotation pipeline consisting of a diffusion model to lift 2D hand keypoint sequences to 4D hand motion. For the forecasting model, we adopt a diffusion loss to account for the multimodality in hand motion distribution. Extensive experiments across 6 datasets show the benefits of training on diverse data with imputed labels (14% improvement) and effectiveness of our lifting (42% better) & forecasting (16.4% gain) models, over the best baselines, especially in zero-shot generalization to everyday images.

cs.CV↗

arXiv2025

Batalin-Fradkin-Vilkovisky quantization and symmetries of FLPR model

We quantize the Friedberg-Lee-Pang-Ren (FLPR) model within the framework of Batalin-Fradkin-Vilkovisky (BFV) formalism. We construct the nilpotent Becchi-Rouet-Stora-Tyutin (BRST) charges using constraints and the fermionic gauge-fixing function by means of admissible gauge conditions. We also derive the BRST invariant effective action (and corresponding symmetries) of the model in both polar and Cartesian coordinates. We demonstrate that the physical states of the system are annihilated by the first-class constraints which is consistent with the Dirac formalism. Moreover, we establish the finite field-dependent BRST (FFBRST) symmetries of the FLPR model. We exhibit the interlink between the BFV-BRST gauge-fixed action and the classical gauge invariant action using FFBRST formulation.

hep-th↗

arXiv2025

Faddeev-Jackiw Approach to Classical Constrained Systems

We accomplish the quantization of a few classical constrained systems \`a la (modified) Faddeev-Jackiw formalism. We analyze the constraint structure and obtain basic brackets of the theory. In addition, we disclose the gauge symmetries within the symplectic framework. We also provide an interpretation for Lagrange multipliers and outline a MATLAB implementation algorithm for symplectic formulation.

hep-th↗

arXiv2025

AlphaOne: Reasoning Models Thinking Slow and Fast at Test Time

This paper presents AlphaOne ($\alpha$1), a universal framework for modulating reasoning progress in large reasoning models (LRMs) at test time. $\alpha$1 first introduces $\alpha$ moment, which represents the scaled thinking phase with a universal parameter $\alpha$. Within this scaled pre-$\alpha$ moment phase, it dynamically schedules slow thinking transitions by modeling the insertion of reasoning transition tokens as a Bernoulli stochastic process. After the $\alpha$ moment, $\alpha$1 deterministically terminates slow thinking with the end-of-thinking token, thereby fostering fast reasoning and efficient answer generation. This approach unifies and generalizes existing monotonic scaling methods by enabling flexible and dense slow-to-fast reasoning modulation. Extensive empirical studies on various challenging benchmarks across mathematical, coding, and scientific domains demonstrate $\alpha$1's superior reasoning capability and efficiency. Project page: https://alphaone-project.github.io/

cs.CL↗

arXiv2025

How Do I Do That? Synthesizing 3D Hand Motion and Contacts for Everyday Interactions

We tackle the novel problem of predicting 3D hand motion and contact maps (or Interaction Trajectories) given a single RGB view, action text, and a 3D contact point on the object as input. Our approach consists of (1) Interaction Codebook: a VQVAE model to learn a latent codebook of hand poses and contact points, effectively tokenizing interaction trajectories, (2) Interaction Predictor: a transformer-decoder module to predict the interaction trajectory from test time inputs by using an indexer module to retrieve a latent affordance from the learned codebook. To train our model, we develop a data engine that extracts 3D hand poses and contact trajectories from the diverse HoloAssist dataset. We evaluate our model on a benchmark that is 2.5-10X larger than existing works, in terms of diversity of objects and interactions observed, and test for generalization of the model across object categories, action categories, tasks, and scenes. Experimental results show the effectiveness of our approach over transformer & diffusion baselines across all settings.

cs.CV↗

arXiv2025

3D Object Reconstruction with mmWave Radars

This paper presents RFconstruct, a framework that enables 3D shape reconstruction using commercial off-the-shelf (COTS) mmWave radars for self-driving scenarios. RFconstruct overcomes radar limitations of low angular resolution, specularity, and sparsity in radar point clouds through a holistic system design that addresses hardware, data processing, and machine learning challenges. The first step is fusing data captured by two radar devices that image orthogonal planes, then performing odometry-aware temporal fusion to generate denser 3D point clouds. RFconstruct then reconstructs 3D shapes of objects using a customized encoder-decoder model that does not require prior knowledge of the object's bound box. The shape reconstruction performance of RFconstruct is compared against 3D models extracted from a depth camera equipped with a LiDAR. We show that RFconstruct can accurately generate 3D shapes of cars, bikes, and pedestrians.

eess.IV↗

arXiv2025

Finding Interest Needle in Popularity Haystack: Improving Retrieval by Modeling Item Exposure

Recommender systems operate in closed feedback loops, where user interactions reinforce popularity bias, leading to over-recommendation of already popular items while under-exposing niche or novel content. Existing bias mitigation methods, such as Inverse Propensity Scoring (IPS) and Off-Policy Correction (OPC), primarily operate at the ranking stage or during training, lacking explicit real-time control over exposure dynamics. In this work, we introduce an exposure-aware retrieval scoring approach, which explicitly models item exposure probability and adjusts retrieval-stage ranking at inference time. Unlike prior work, this method decouples exposure effects from engagement likelihood, enabling controlled trade-offs between fairness and engagement in large-scale recommendation platforms. We validate our approach through online A/B experiments in a real-world video recommendation system, demonstrating a 25% increase in uniquely retrieved items and a 40% reduction in the dominance of over-popular content, all while maintaining overall user engagement levels. Our results establish a scalable, deployable solution for mitigating popularity bias at the retrieval stage, offering a new paradigm for bias-aware personalization.

cs.IR↗