arXiv subjects

Shurui Li

Publications and source records attributed to Shurui Li.

18 recordsLinked to original sources

arXiv2026

Qwen-RobotNav Technical Report: A Scalable Navigation Model Designed for an Agentic Navigation System

Agentic navigation systems require a base navigation model whose observation strategy can be externally reconfigured at inference time, because instruction following, object search, target tracking, and autonomous driving share the same perception-planning backbone yet demand fundamentally different strategies for consuming the visual stream. We present Qwen-RobotNav, a scalable navigation model built on Qwen-RobotNav that addresses it through a parameterised interface with two complementary dimensions: multiple task modes that select the navigation behaviour, and controllable observation parameters (e.g., token budget, per-camera weights) that govern how visual history is encoded. With training-time randomization over all parameters, Qwen-RobotNav is robust to any inference-time configuration requiring zero architectural modification to the Qwen-RobotNav backbone. We train Qwen-RobotNav on 15.6M samples; co-training with vision-language data prevents the collapse into reactive action-sequence mappers observed in trajectory-only training. The parameterised interface also makes Qwen-RobotNav a natural building block for agentic systems: for long-horizon scenarios, an upper-level planner decomposes goals into sub-tasks and dynamically switches Qwen-RobotNav's task mode and context strategy mid-episode, composing complex behaviours from repeated calls to the same model. Extensive experiments show that Qwen-RobotNav sets new state-of-the-art results across major navigation benchmarks. The model exhibits favourable scaling from 2B to 8B parameters, with joint multi-task training developing a shared spatial-planning substrate that transfers across task families, and demonstrates strong zero-shot generalisation to real-world robots across diverse environments.

cs.RO↗

arXiv2026

Recovering Policy-Induced Errors: Benchmarking and Trajectory Synthesis for Robust GUI Agents

While GUI agents have advanced rapidly, they often lack the robustness to recover from their own errors, hindering real-world deployment. To bridge this gap at both the evaluation and data levels, we introduce GUI-RobustEval and propose Robustness-driven Trajectory Synthesis. GUI-RobustEval contains $1,216$ executable test cases that systematically measure error recovery capabilities across a broad and realistic spectrum of error modes. At the data level, RoTS is a scalable synthesis framework that creates $800k$ high-quality data via a tree-based pipeline that proactively discovers diverse error modes and synthesizes corresponding recovery steps. Our two models, RoTS-7B and RoTS-32B, fine-tuned on our dataset, both demonstrate significant gains on GUI-RobustEval and traditional GUI benchmarks. Notably, RoTS-32B achieves state-of-the-art performance on OSWorld, with a $47.4\%$ success rate and a $33.8\%$ All-Pass@4 score, suggesting that improved long-horizon error recovery ability contributes to both robustness and overall performance. Our code is available at https://github.com/AlibabaResearch/RoTS.

cs.CV↗

arXiv2026

Long Live The Balance: Information Bottleneck Driven Tree-based Policy Optimization

Recent advances in online reinforcement learning (RL) for large language models (LLMs) have demonstrated promising performance in complex reasoning tasks. However, they often exhibit an imbalanced exploration-exploitation trade-off, resulting in unstable optimization and sub-optimal performance. We introduce IB-Score, a novel metric grounded in Information Bottleneck theory that evaluates policy's exploration-exploitation balance by quantifying the trade-off between step-level reasoning diversity and mutual information shared with the correct answer. Analysis based on IB-Score shows that popular online RL approaches (e.g., GRPO) with common regularizers fail to consistently maintain balance during training with suboptimal results. To address this, we propose Information Bottleneck-driven Tree-based Policy Optimization (IB-TPO), a principled framework that formulates IB-Score as a fine-grained optimization objective and utilizes a novel IB-guided tree sampling strategy that not only improves the efficiency of online sampling with 50% more trajectories under the same token budget, but also reuses the tree structure for effective IB-Score Monte Carlo estimation. Extensive experiments across standard benchmarks show that our method significantly outperforms GRPO baseline by 2.9% to 3.6% and also outperforms other state-of-the-art online RL approaches. Our code is available at https://github.com/alibaba/EfficientRL.

cs.LG↗

arXiv2026

One Brain, Omni Modalities: Towards Unified Non-Invasive Brain Decoding with Large Language Models

Deciphering brain function through non-invasive recordings requires synthesizing complementary high-frequency electromagnetic (EEG/MEG) and low-frequency metabolic (fMRI) signals. However, despite their shared neural origins, extreme discrepancies have traditionally confined these modalities to isolated analysis pipelines, hindering a holistic interpretation of brain activity. To bridge this fragmentation, we introduce \textbf{NOBEL}, a \textbf{n}euro-\textbf{o}mni-modal \textbf{b}rain-\textbf{e}ncoding \textbf{l}arge language model (LLM) that unifies these heterogeneous signals within the LLM's semantic embedding space. Our architecture integrates a unified encoder for EEG and MEG with a novel dual-path strategy for fMRI, aligning non-invasive brain signals and external sensory stimuli into a shared token space, then leverages an LLM as a universal backbone. Extensive evaluations demonstrate that NOBEL serves as a robust generalist across standard single-modal tasks. We also show that the synergistic fusion of electromagnetic and metabolic signals yields higher decoding accuracy than unimodal baselines, validating the complementary nature of multiple neural modalities. Furthermore, NOBEL exhibits strong capabilities in stimulus-aware decoding, effectively interpreting visual semantics from multi-subject fMRI data on the NSD and HAD datasets while uniquely leveraging direct stimulus inputs to verify causal links between sensory signals and neural responses. NOBEL thus takes a step towards unifying non-invasive brain decoding, demonstrating the promising potential of omni-modal brain understanding.

q-bio.NC↗

arXiv2025

Perfect AI Mimicry and the Epistemology of Consciousness: A Solipsistic Dilemma

Rapid advances in artificial intelligence necessitate a re-examination of the epistemological foundations upon which we attribute consciousness. As AI systems increasingly mimic human behavior and interaction with high fidelity, the concept of a "perfect mimic"-an entity empirically indistinguishable from a human through observation and interaction-shifts from hypothetical to technologically plausible. This paper argues that such developments pose a fundamental challenge to the consistency of our mind-recognition practices. Consciousness attributions rely heavily, if not exclusively, on empirical evidence derived from behavior and interaction. If a perfect mimic provides evidence identical to that of humans, any refusal to grant it equivalent epistemic status must invoke inaccessible factors, such as qualia, substrate requirements, or origin. Selectively invoking such factors risks a debilitating dilemma: either we undermine the rational basis for attributing consciousness to others (epistemological solipsism), or we accept inconsistent reasoning. I contend that epistemic consistency demands we ascribe the same status to empirically indistinguishable entities, regardless of metaphysical assumptions. The perfect mimic thus acts as an epistemic mirror, forcing critical reflection on the assumptions underlying intersubjective recognition in light of advancing AI. This analysis carries significant implications for theories of consciousness and ethical frameworks concerning artificial agents.

cs.AI↗

arXiv2025

Modeling the language cortex with form-independent and enriched representations of sentence meaning reveals remarkable semantic abstractness

The human language system represents both linguistic forms and meanings, but the abstractness of the meaning representations remains debated. Here, we searched for abstract representations of meaning in the language cortex by modeling neural responses to sentences using representations from vision and language models. When we generate images corresponding to sentences and extract vision model embeddings, we find that aggregating across multiple generated images yields increasingly accurate predictions of language cortex responses, sometimes rivaling large language models. Similarly, averaging embeddings across multiple paraphrases of a sentence improves prediction accuracy compared to any single paraphrase. Enriching paraphrases with contextual details that may be implicit (e.g., augmenting "I had a pancake" to include details like "maple syrup") further increases prediction accuracy, even surpassing predictions based on the embedding of the original sentence, suggesting that the language system maintains richer and broader semantic representations than language models. Together, these results demonstrate the existence of highly abstract, form-independent meaning representations within the language cortex.

cs.CL↗

arXiv2025

Near-energy-free Photonic Fourier Transformation for Convolution Operation Acceler

Convolutional operations are computationally intensive in artificial intelligence services, and their overhead in electronic hardware limits machine learning scaling. Here, we introduce a photonic joint transform correlator (pJTC) using a near-energy-free on-chip Fourier transformation to accelerate convolution operations. The pJTC reduces computational complexity for both convolution and cross-correlation from O(N4) to O(N2), where N2 is the input data size. Demonstrating functional Fourier transforms and convolution, this pJTC achieves 98.0% accuracy on an exemplary MNIST inference task. Furthermore, a wavelength-multiplexed pJTC architecture shows potential for high throughput and energy efficiency, reaching 305 TOPS/W and 40.2 TOPS/mm2, based on currently available foundry processes. An efficient, compact, and low-latency convolution accelerator promises to advance next-generation AI capabilities across edge demands, high-performance computing, and cloud services.

physics.optics↗

arXiv2025

CIMPool: Scalable Neural Network Acceleration for Compute-In-Memory using Weight Pools

Compute-in-memory (CIM) based neural network accelerators offer a promising solution to the Von Neumann bottleneck by computing directly within memory arrays. However, SRAM CIM faces limitations in executing larger models due to its cell size and on-chip memory constraints. This work proposes CIMPool, a CIM-aware compression and acceleration framework that counters this limitation through a weight sharing-based compression technique, aptly named `Weight Pool,' enabling significantly larger neural networks to be accommodated within on-chip memory constraints. This method minimizes the accuracy trade-off typically associated with parameter compression, allowing CIMPool to achieve a significantly larger compression ratio compared to the traditional quantization method with iso-accuracy. Furthermore, CIMPool co-optimizes the compression algorithm, hardware, and dataflow to efficiently implement the hardware permutation required by weight pool compression, with negligible area and throughput overhead. Empirical results demonstrate that CIMPool can achieve 8-bit level accuracy with an effective 0.5-bit precision, reduce chip area by 62.3% for ResNet-18, and enable the execution of an order of magnitude larger models for a given area budget in SRAM CIMs. When DRAM is used to store weights, CIMPool can reduce the total energy by 3.24x compared to iso-accuracy traditional CIMs.

cs.AR↗

arXiv2023

Training Neural Networks for Execution on Approximate Hardware

Approximate computing methods have shown great potential for deep learning. Due to the reduced hardware costs, these methods are especially suitable for inference tasks on battery-operated devices that are constrained by their power budget. However, approximate computing hasn't reached its full potential due to the lack of work on training methods. In this work, we discuss training methods for approximate hardware. We demonstrate how training needs to be specialized for approximate hardware, and propose methods to speed up the training process by up to 18X.

cs.LG↗

arXiv2022

PhotoFourier: A Photonic Joint Transform Correlator-Based Neural Network Accelerator

The last few years have seen a lot of work to address the challenge of low-latency and high-throughput convolutional neural network inference. Integrated photonics has the potential to dramatically accelerate neural networks because of its low-latency nature. Combined with the concept of Joint Transform Correlator (JTC), the computationally expensive convolution functions can be computed instantaneously (time of flight of light) with almost no cost. This 'free' convolution computation provides the theoretical basis of the proposed PhotoFourier JTC-based CNN accelerator. PhotoFourier addresses a myriad of challenges posed by on-chip photonic computing in the Fourier domain including 1D lenses and high-cost optoelectronic conversions. The proposed PhotoFourier accelerator achieves more than 28X better energy-delay product compared to state-of-art photonic neural network accelerators.

cs.AR↗

arXiv2022

Bit-serial Weight Pools: Compression and Arbitrary Precision Execution of Neural Networks on Resource Constrained Processors

Applications of neural networks on edge systems have proliferated in recent years but the ever-increasing model size makes neural networks not able to deploy on resource-constrained microcontrollers efficiently. We propose bit-serial weight pools, an end-to-end framework that includes network compression and acceleration of arbitrary sub-byte precision. The framework can achieve up to 8x compression compared to 8-bit networks by sharing a pool of weights across the entire network. We further propose a bit-serial lookup based software implementation that allows runtime-bitwidth tradeoff and is able to achieve more than 2.8x speedup and 7.5x storage compression compared to 8-bit weight pool networks, with less than 1% accuracy drop.

cs.LG↗

arXiv2021

High Throughput Multi-Channel Parallelized Diffraction Convolutional Neural Network Accelerator

Convolutional neural networks are paramount in image and signal processing including the relevant classification and training tasks alike and constitute for the majority of machine learning compute demand today. With convolution operations being computationally intensive, next generation hardware accelerators need to offer parallelization and algorithmic-hardware homomorphism. Fortunately, diffractive display optics is capable of million-channel parallel data processing at low latency, however, thus far only showed tens of Hertz slow single image and kernel capability, thereby significantly underdelivering from its performance potential. Here, we demonstrate an operation-parallelized high-throughput Fourier optic convolutional neural network accelerator. For the first time simultaneously processing of multiple kernels in Fourier domain enabled by optical diffraction has been achieved alongside with already conventional in the field input parallelism. Additionally, we show an about one hundred times system speed up over existing optical diffraction-based processors and this demonstration rivals performance of modern electronic solutions. Therefore, this system is capable of processing large-scale matrices about ten times faster than state of art electronic systems.

cs.LG↗

arXiv2021

Optimal model for fewer-qubit CNOT gates with Rydberg atoms

Fewer-qubit quantum logic gate, serving as a basic unit for constructing universal multiqubit gates, has been widely applied in quantum computing and quantum information. However, traditional constructions for fewer-qubit gates often utilize a multi-pulse protocol which inevitably suffers from serious intrinsic errors during the gate execution. In this article, we report an optimal model about universal two- and three-qubit CNOT gates mediated by excitation to Rydberg states with easily-accessible van der Waals interactions. This gate depends on a global optimization to implement amplitude and phase modulated pulses via genetic algorithm, which can facilitate the gate operation with fewer optical pulses. Compared to conventional multi-pulse piecewise schemes, our gate can be realized by simultaneous excitation of atoms to the Rydberg states, saving the time for multi-pulse switching at different spatial locations. Our numerical simulations show that a single-pulse two(three)-qubit CNOT gate is possibly achieved with a fidelity of 99.23$\%$(90.39$\%$) for two qubits separated by 7.10 $\mu$m when the fluctuation of Rydberg interactions is excluded. Our work is promising for achieving fast and convenient multiqubit quantum computing in the study of neutral-atom quantum technology.

quant-ph↗

arXiv2021

Revisiting the dynamics of Bose-Einstein condensates in a double well by deep learning with a hybrid network

Deep learning, accounting for the use of an elaborate neural network, has recently been developed as an efficient and powerful tool to solve diverse problems in physics and other sciences. In the present work, we propose a novel learning method based on a hybrid network integrating two different kinds of neural networks: Long Short-Term Memory(LSTM) and Deep Residual Network(ResNet), in order to overcome the difficulty met in numerically simulating strongly-oscillating dynamical evolutions of physical systems. By taking the dynamics of Bose-Einstein condensates in a double-well potential as an example, we show that our new method makes a high efficient pre-learning and a high-fidelity prediction about the whole dynamics. This benefits from the advantage of the combination of the LSTM and the ResNet and is impossibly achieved by a single network in the case of direct learning. Our method can be applied for simulating complex cooperative dynamics in a system with fast multiple-frequency oscillations with the aid of auxiliary spectrum analysis.

physics.comp-ph↗

arXiv2021

SWIS -- Shared Weight bIt Sparsity for Efficient Neural Network Acceleration

Quantization is spearheading the increase in performance and efficiency of neural network computing systems making headway into commodity hardware. We present SWIS - Shared Weight bIt Sparsity, a quantization framework for efficient neural network inference acceleration delivering improved performance and storage compression through an offline weight decomposition and scheduling algorithm. SWIS can achieve up to 54.3% (19.8%) point accuracy improvement compared to weight truncation when quantizing MobileNet-v2 to 4 (2) bits post-training (with retraining) showing the strength of leveraging shared bit-sparsity in weights. SWIS accelerator gives up to 6x speedup and 1.9x energy improvement overstate of the art bit-serial architectures.

cs.LG↗

arXiv2020

Channel Tiling for Improved Performance and Accuracy of Optical Neural Network Accelerators

Low latency, high throughput inference on Convolution Neural Networks (CNNs) remains a challenge, especially for applications requiring large input or large kernel sizes. 4F optics provides a solution to accelerate CNNs by converting convolutions into Fourier-domain point-wise multiplications that are computationally 'free' in optical domain. However, existing 4F CNN systems suffer from the all-positive sensor readout issue which makes the implementation of a multi-channel, multi-layer CNN not scalable or even impractical. In this paper we propose a simple channel tiling scheme for 4F CNN systems that utilizes the high resolution of 4F system to perform channel summation inherently in optical domain before sensor detection, so the outputs of different channels can be correctly accumulated. Compared to state of the art, channel tiling gives similar accuracy, significantly better robustness to sensing quantization (33\% improvement in required sensing precision) error and noise (10dB reduction in tolerable sensing noise), 0.5X total filters required, 10-50X+ throughput improvement and as much as 3X reduction in required output camera resolution/bandwidth. Not requiring any additional optical hardware, the proposed channel tiling approach addresses an important throughput and precision bottleneck of high-speed, massively-parallel optical 4F computing systems.

cs.ET↗

arXiv2020

Massively Parallel Amplitude-Only Fourier Neural Network

Machine-intelligence has become a driving factor in modern society. However, its demand outpaces the underlying electronic technology due to limitations given by fundamental physics such as capacitive charging of wires, but also by system architecture of storing and handling data, both driving recent trends towards processor heterogeneity. Here we introduce a novel amplitude-only Fourier-optical processor paradigm capable of processing large-scale ~(1,000 x 1,000) matrices in a single time-step and 100 microsecond-short latency. Conceptually, the information-flow direction is orthogonal to the two-dimensional programmable-network, which leverages 10^6-parallel channels of display technology, and enables a prototype demonstration performing convolutions as pixel-wise multiplications in the Fourier domain reaching peta operations per second throughputs. The required real-to-Fourier domain transformations are performed passively by optical lenses at zero-static power. We exemplary realize a convolutional neural network (CNN) performing classification tasks on 2-Megapixel large matrices at 10 kHz rates, which latency-outperforms current GPU and phase-based display technology by one and two orders of magnitude, respectively. Training this optical convolutional layer on image classification tasks and utilizing it in a hybrid optical-electronic CNN, shows classification accuracy of 98% (MNIST) and 54% (CIFAR-10). Interestingly, the amplitude-only CNN is inherently robust against coherence noise in contrast to phase-based paradigms and features an over 2 orders of magnitude lower delay than liquid crystal-based systems. Beyond contributing to novel accelerator technology, scientifically this amplitude-only massively-parallel optical compute-paradigm can be far-reaching as it de-validates the assumption that phase-information outweighs amplitude in optical processors for machine-intelligence.

eess.IV↗

arXiv2019

You May Not Need Order in Time Series Forecasting

Time series forecasting with limited data is a challenging yet critical task. While transformers have achieved outstanding performances in time series forecasting, they often require many training samples due to the large number of trainable parameters. In this paper, we propose a training technique for transformers that prepares the training windows through random sampling. As input time steps need not be consecutive, the number of distinct samples increases from linearly to combinatorially many. By breaking the temporal order, this technique also helps transformers to capture dependencies among time steps in finer granularity. We achieve competitive results compared to the state-of-the-art on real-world datasets.

cs.LG↗