Researcher profile

Yuqi Li

Yuqi Li contributes to research discovery and scholarly infrastructure.

ResearcherAffiliation not importedOpen to collaborate

Artificial Intelligence Computer Vision physics.optics Robotics

Trust snapshot

Quick read

Trust 15 - UnverifiedVerification L1Unclaimed author

3works

0followers

4topics

4close collaborators

Actions

Decide how to stay connected

Follow researcher0

Claiming links this public author record to a researcher profile and unlocks direct collaboration workflows.

Direct collaboration

Open a focused conversation when the fit is right

Claim this author entity first to unlock direct invitations.

Inspect adjacent work, topics, institutions and collaborators without jumping out to a separate graph page.

Building this graph slice

BZPEER is loading the nearby papers, people, topics and institutions for this page.

preprint2026arXiv

First Thin-Film Lithium Tantalate Polarization Controller Enabling Reset-Free Mrad/s Tracking for Optical Interconnects

The rapid escalation of computing power driven by large-scale artificial intelligence is placing unprecedented demands on the bandwidth, latency, and energy efficiency of data-center interconnects (DCIs). Self-homodyne coherent (SHC) transmission is a promising architecture because it preserves the spectral efficiency of coherent detection while greatly simplifying digital signal processing, but its practical deployment is critically limited by random and often ultrafast state-of-polarization (SOP) fluctuations that induce carrier fading and destabilize coherent reception. Here we report the first integrated polarization controller based on thin-film lithium tantalate (TFLT), enabling reset-free polarization tracking at Mrad/s speeds. The four-stage electro-optic device exhibits polarization-dependent loss (PDL) below 0.3 dB, a half-wave voltage below 2.5 V, high modulation bandwidth, and negligible DC drift. To accommodate the finite tuning range of integrated phase shifters, we develop a finite-boundary gradient-descent (FBGD) control algorithm that ensures reset-free SOP evolution with no phase jump. The implemented adaptive polarization controller (APC) is validated through both standalone polarization-tracking measurements and a dual-polarization 16-QAM SHC 400-Gbps transmission system. Transient polarization disturbances can be tracked at speeds up to 2 Mrad/s, while stable reset-free operation under continuous polarization disturbances is maintained up to 1 Mrad/s. This reset-free performance represents more than doubling the state of the art, while the pre-FEC bit-error rates remain below the HD-FEC threshold under realistic DCI conditions and lightning-scale polarization disturbances. These results establish TFLT as a new platform for ultrafast, low-power, reset-free, and drift-free polarization control in coherent optical interconnects and beyond.

preprint2026arXiv

RoboAlign-R1: Distilled Multimodal Reward Alignment for Robot Video World Models

Existing robot video world models are typically trained with low-level objectives such as reconstruction and perceptual similarity, which are poorly aligned with the capabilities that matter most for robot decision making, including instruction following, manipulation success, and physical plausibility. They also suffer from error accumulation in long-horizon autoregressive prediction. We present RoboAlign-R1, a framework that combines reward-aligned post-training with stabilized long-horizon inference for robot video world models. We construct RobotWorldBench, a benchmark of 10,000 annotated video-instruction pairs collected from four robot data sources, and train a multimodal teacher judge, RoboAlign-Judge, to provide fine-grained six-dimensional evaluation of generated videos. We then distill the teacher into a lightweight student reward model for efficient reinforcement-learning-based post-training. To reduce long-horizon rollout drift, we further introduce Sliding Window Re-encoding (SWR), a training-free inference strategy that periodically refreshes the generation context. Under our in-domain evaluation protocol, RoboAlign-R1 improves the aggregate six-dimension score by 10.1% over the strongest baseline, including gains of 7.5% on Manipulation Accuracy and 4.6% on Instruction Following; these ranking improvements are further supported by an external VLM-based cross-check and a blinded human study. Meanwhile, SWR improves long-horizon prediction quality with only about 1% additional latency, yielding a 2.8% gain in SSIM and a 9.8% reduction in LPIPS. Together, these results show that reward-aligned post-training and stabilized long-horizon decoding improve task consistency, physical realism, and long-horizon prediction quality in robot video world models.

preprint2026arXiv

Self-Evolving Spatial Reasoning in Vision Language Models via Geometric Logic Consistency

Vision-Language Models (VLMs) have made striking progress, yet their spatial reasoning remains fragile: models that answer an original input correctly can still fail under paired transformations with predictable answer mappings, revealing a gap between instance-level correctness and robust spatial reasoning. To address this, we propose Spatial Alignment via Geometric Evolution (SAGE), a self-evolving framework that enforces logical consistency in VLMs through geometric and linguistic duality operations. SAGE incorporates duality consistency as an auxiliary reward within GRPO training, encouraging models to produce logically coherent answers across original and transformed inputs. A dynamic operation pool continuously probes for inconsistencies, promoting challenging operations and retiring mastered ones, so that training focuses on the most informative signals. SAGE is model-agnostic, data-efficient compared to prior GRPO methods, and can be applied as a lightweight post-training stage to any existing VLM. Experiments on video and spatial reasoning benchmarks demonstrate consistent improvements over strong baselines and enhanced generalization to unseen data.

Institution

Affiliation not imported yet

This author record came from a source that does not expose affiliation metadata. Once the author claims the profile or we enrich the record from another provider, this section will link to the concrete institution.

Topic footprint