arXiv subjects

Li Ding

Publications and source records attributed to Li Ding.

At least 19 recordsLinked to original sources

arXiv2026

From Visual Affinity to Learning Credit in Multimodal Reinforcement Learning

Reinforcement learning with verifiable rewards (RLVR) improves multimodal reasoning, but uniformly assigning a single response-level advantage to all tokens overlooks the distinct roles of visual observations and subsequent reasoning. Visual affinity alone is insufficient: predictive states may align with both image and prompt-context representations, and upweighting valid visual observations can amplify their negative-advantage updates when later reasoning fails. To address this issue, we propose \textbf{Outcome-Conditioned Visual Credit (OVC)}, a plug-in framework that separates visual-association measurement from token-level credit assignment and is adaptable to policy-optimization methods with uniform response-level advantage allocation. Relative Visual Encoding constructs a context-relative visual coordinate by contrasting each predictive state's affinity to image and prompt-context representations. Outcome-conditioned Evidence Calibration combines this coordinate with uncertainty-scaled predictive features and calibrates their contributions to outcome-discriminative evidence through mean contrasts between positive- and negative-advantage responses, scaled by regularized within-side variation. Visual association thus informs credit through observed outcome differences rather than directly determining update priority. Normalized Credit Allocation conditions the calibrated evidence on the response's advantage sign to produce positive token weights with unit global mean. These detached weights preserve advantage signs while modulating individual token-gradient contributions. OVC reuses existing rollouts without auxiliary reward models, process annotations, or inference-time modules. Instantiated through GSPO-token, OVC retains sequence-level clipping and consistently improves six-benchmark mean accuracy over GSPO across three models, with gains of 1.86-2.39 percentage points.

cs.AI↗

arXiv2026

Nemotron 3.5 Bolt 74B-A9B 24GB: A Hybrid Vision-Language Model for Local AI

We introduce Nemotron 3.5 Bolt 74B-A9B 24GB, a hybrid Mamba-Transformer vision-language model that optimizes speed and intelligence per GB of memory used for local agentic workloads. Derived from the Nemotron 3.5 Super VL 120B-A12B model family, Nemotron 3.5 Bolt 74B-A9B 24GB combines Star Elastic compression with 2-bit Quantization-Aware Distillation for memory-constrained deployment. The model is optimized for reasoning, agentic tool use, and web & terminal interactions. We also trained a LiLiCorr speculative drafter that is released alongside the model to reach our target inference speed. The resulting model achieves a higher accuracy than leading models with similar memory footprints and speed across many benchmarks for general agentic capabilities (e.g., GDPVal v2), web browsing (e.g., BrowseComp), and terminal use (e.g., Terminal-Bench 2.1), with a relatively high TPS of 84 on RTX Spark that is comparable to models with much lower active parameter counts and more than sufficient for long-horizon agentic applications.

cs.CV↗

arXiv2026

Bongard: Training Machine Intuition

Human intelligence relies heavily on learned intuition: recognising patterns and judging situations without explicitly unfolding every intermediate step. We introduce Bongard, an open-weight System One model that treats machine intuition as an independent capability to design and train. A T5Gemma 2 4B-4B encoder-decoder separates reading the evidence from making judgments. The encoder reads the state bidirectionally together with the question instructions, and separate decoder branches share this encoding, so many judgments about the same situation require only one reading of the state. A trained head returns probabilities over the supplied candidates without generating text. Training proceeds in three stages, from supervised judgments to semantic relationships to action outcomes, and each stage updates all 7.09 billion trainable parameters on one Blackwell GPU. Joint-embedding post-training raises accuracy on held-out rephrasings from 75.7% to 85.9%. A sandbox stage then learns outcome distributions from action rollouts and exact oracles, raising accuracy on a frozen sandbox panel from 50.6% to 64.8%. On DecisionBench, the final model reaches 78.05% accuracy over 23,900 decisions and ranks fourth of 61 systems in the public comparison. On one RTX PRO 6000, its median latency is 36 ms for short requests, and 32 questions about one state take 221 ms. Bongard demonstrates that machine intuition can be systematically trained via representation learning and outcome feedback, providing an open, efficient alternative for high-throughput decision workloads.

cs.CL↗

arXiv2026

LoRA-generating hypernetworks for efficient on-device LLM generative personalization

On-device large language models (`LLMs'), e.g. running on mobile phones, are ripe for improvement via personalization. The limited compute resources of mobile devices impose limits on model scale and thus model quality, making any realizable quality gains highly impactful. At the same time, their personal nature (i.e., the close coupling to a particular user) means that a given on-device LLM tends to be used in similar, predictable patterns over the course of time. This paper presents a novel method for personalizing on-device LLMs. It trains a hypernetwork to map a user's context tokens to a low-rank adaptation (`LoRA') well-suited to that user. Once the trained common artifacts are deployed to users' devices, each user uses the hypernetwork to synthesize (entirely on device) a personalized LoRA. This approach blends the benefits while avoiding the drawbacks of two existing approaches to LLM customization: in-context learning (`ICL') and parameter-efficient fine-tuning (`PEFT'). Like ICL (and unlike PEFT), the on-device phase of our approach is computationally feasible, requiring only forward passes through neural networks. Like PEFT (and unlike ICL), our approach modifies the `target' base LLM via weights (the LoRA), avoiding negative consequences (e.g. increased latency) associated with extending the input sequence. Our approach is particularly well-suited to the mobile device regime. Apart from the on-device compute and latency benefits mentioned, it also requires minimal additional storage, as internally its architecture partly leverages the same LLM weights as belong to the target LLM to be personalized. We demonstrate the benefits of LoRA-generating hypernetworks on several representative personalization datasets, comparing against baselines like ICL and PEFT. Of note, our personalization experiments focus on more challenging and less studied long-form text generation tasks.

cs.LG↗

arXiv2026

Stochastic Corridor Time Network Capacity Planning for Low Altitude Airspace Systems

Regulators in China, the United States, and the European Union now provide low-altitude airspace access as priced, time-windowed corridor authorizations, booked in advance and forfeited if unused. We ask how much capacity a UAV logistics planner should reserve on each corridor--time unit before demand is realized, to maximize expected profit net of reservation cost. Reserved capacity cannot be transferred across corridors or time windows and is consumed jointly along time-respecting paths, so reservations are coupled through the network in ways that models with exogenous airspace capacity cannot capture. We formulate a two-stage stochastic program whose recourse selects and routes accepted requests on a time-expanded network, prove its arc-based and path-packing forms equivalent, and solve it by Benders decomposition with column-generated subproblems. The decomposition operates on the LP relaxation, and all reported reservation and routing decisions are recovered as integer plans. Computational experiments achieve single-digit LP-Benders gaps on moderate-sized networks and extend to much larger instances through a truncated-path approach. A Shenzhen case study shows reservations concentrating on structurally central corridors, with demand level and reservation price having more influence on the quantity of capacity reserved than the selection of corridors.

cs.CE↗

arXiv2026

A High-Precision Frequency Locking Method Based on All-Phase FFT Demonstrated on a Crystal Oscillator with Rubidium Clock Reference

This article proposes a novel frequency-locking method based on frequency-domain unbiased phase estimation (FDUPE) for high-precision frequency control. By performing weighted recombination of the acquired data followed by Fourier-transform processing, the phase at the center of the data segment can be estimated without bias, making the method suitable for frequency-locking applications. The principle of the proposed method is analyzed, and an electronic prototype is developed to experimentally validate its feasibility. In the prototype, analog-to-digital converters (ADCs) are used for signal digitization, and a field-programmable gate array (FPGA) is used to implement the FDUPE algorithm. A digital proportional-integral-derivative (PID) controller is also implemented on the FPGA to provide feedback for accurate frequency locking. In the experiment, a (10~\mathrm{MHz}) voltage-controlled oscillator (VCO) with a free-running Allan deviation of (1 \times 10^{-9}) at (1~\mathrm{s}) is used as the device under test (DUT), while a rubidium atomic clock with an Allan deviation of (2 \times 10^{-11}) at (1~\mathrm{s}) serves as the high-stability reference source. Experimental results show that the proposed system achieves excellent locking performance, reducing the standard deviation of frequency fluctuations from (12.75~\mathrm{mHz}) root-mean-square (rms) in the free-running state to (0.88~μ\mathrm{Hz}) rms after locking. Correspondingly, the Allan deviation at (10~\mathrm{s}) is reduced from (9.6 \times 10^{-10}) to (1.45 \times 10^{-14}), representing a five-order-of-magnitude improvement in frequency stability.

physics.ins-det↗

arXiv2026

Heterogeneous Parallelism for Multimodal Large Language Model Training

Foundation model training is becoming multimodal, from post-training pipelines to large-scale pretraining. As modality coverage broadens, context windows grow, and encoder LLM scales diverge, a single LLM-centric TP/CP/PP/DP/EP layout increasingly limits throughput. This coupling forces encoders to inherit LLM-driven sharding and placement choices that can add communication, limit encoder parallelism, or constrain the LLM schedule; the mismatch is most pronounced at long contexts, where LLM context parallelism is needed for the fused multimodal sequence but encoder inputs remain bounded. We present heterogeneous parallelism for multimodal large language model training, an abstraction that lets modules in one end-to-end graph use independent layouts and rank placements, supporting colocated execution on shared GPUs and non-colocated execution on disjoint rank sets. The key challenge is preserving boundary tensor semantics across independent layouts: forward activations must be materialized for the destination layout, while backward gradients must be routed back to the source layout. We address this with boundary communicators that implement forward and backward layout transforms, plus scheduling extensions for both placement modes. We evaluate optimized homogeneous, colocated heterogeneous, and non-colocated heterogeneous configurations across multimodal workloads and GPU scales to characterize when added layout and placement freedom exposes a better operating point. Across this sweep, colocated heterogeneity improves TFLOPS/GPU by up to 49.3%, while non-colocated heterogeneity improves aggregate token throughput by up to 13.0% and TFLOPS/GPU by up to 9.6%. We validate loss convergence parity against homogeneous baselines and release the system as an open-source Megatron-LM extension.

cs.LG↗

arXiv2026

High-Density Horizontal Arrays of Single-Chirality Carbon Nanotubes

Highly ordered high-density arrays of single-chirality single-walled carbon nanotubes (SWCNTs) are greatly desired for exploring the intrinsic anisotropic properties and collective performance of such 1-dimensional (1D) nanomaterials. Here we present a Marangoni flow-induced self-assembly (MISA) strategy to fabricate monolayered SWCNT arrays achieving a packing density of ~200 ${μm}^{-1}$ and a 2-dimensional order parameter ($S_{2\mathrm{D}}$) of ~0.95. Relying on its general compatibility with both organic and aqueous dispersions, we prepare single-chirality and enantiomer-pure SWCNT arrays from organic and aqueous dispersions resulting from the sorting processes. The anisotropic optical and electrical properties of the arrays are demonstrated by the polarization-dependent Rabi splitting as well as polarized near-infrared light emission and detection. With the great tolerance to solutions, substrates, and materials, as well as the feasibility and controllability, MISA shows great potential in the assembly of 1D nanomaterials.

cond-mat.mtrl-sci↗

arXiv2026

Unlocking Biological Workflows for Robust Protein-Text Question Answering: A Dual-Dimensional RAG Framework

Protein-Text Question Answering (QA) is crucial for interpreting biological sequences through natural language. The integration of Large Language Models (LLMs) with Retrieval-Augmented Generation (RAG) that efficiently leverages biological databases and facilitates reasoning offers a potent approach for it. However, constrained by the standard RAG pipeline, these models often rely on curated, static datasets instead of expert-proven biological workflows, lacking the fine-grained information processing and struggling to generalize to novel (OOD) proteins. To bridge this gap, we propose 2D-ProteinRAG, a novel framework that empowers LLMs to operate within the gold-standard biological research workflow (BLAST). To further extract high-quality information from noisy retrieval contexts, we introduce a dual-dimensional (2D) filtering strategy following the expert analytical paradigms. Horizontal Fine-grained Attribute Alignment utilizes a lightweight, intent-aware discriminative filter to prune irrelevant metadata and align database entries with specific user queries. Vertical Homology-based Semantic Denoising resolves functional contradictions and redundancy across multiple homologs via hierarchical clustering. Extensive evaluations on both In-Distribution and diverse biological OOD benchmarks demonstrate that 2D-ProteinRAG consistently achieves state-of-the-art performance, outperforming fine-tuned baselines and other RAG methods. Our results validate the framework's robustness and scalability, providing a practical solution for interpreting protein functions in real-world scientific scenarios.

cs.IR↗

arXiv2026

Nemotron 3 Nano Omni: Efficient and Open Multimodal Intelligence

We introduce Nemotron 3 Nano Omni, the latest model in the Nemotron multimodal series and the first to natively support audio inputs alongside text, images, and video. Nemotron 3 Nano Omni delivers consistent accuracy improvements over its predecessor, Nemotron Nano V2 VL, across all modalities, enabled by advances in architecture, training data and recipes. In particular, Nemotron 3 delivers leading results in real-world document understanding, long audio-video comprehension, and agentic computer use. Built on the highly efficient Nemotron 3 Nano 30B-A3B backbone, Nemotron 3 Nano Omni further incorporates innovative multimodal token-reduction techniques to deliver substantially lower inference latency and higher throughput than other models of similar size. We are releasing model checkpoints in BF16, FP8, and FP4 formats, along with portions of the training data and codebase to facilitate further research and development.

cs.LG↗

arXiv2026

Advancing Edge Classification through High-Dimensional Causal Modeling of Node-Edge Interplay

Edge classification, a crucial task for graph applications, remains relatively under-explored compared to link prediction. Current methods often overlook the potential causal influences of node features on edge features, leading to a loss of relevant prior information. In this work, we present an empirical exploration using the Causal Edge Classification Framework (CECF). Unlike conventional causal inference methods, CECF is the first framework to apply causal inference principles to the edge classification task and to explore modeling edge features as a high-dimensional treatment within a causal framework. Based on the node embedding of Graph Neural Network (GNN), CECF seeks to learn a balanced representation of high-dimensional edge features by mitigating the potential influence of node features. Then, a cross-attention network captures the complex dependencies between node and edge features for final edge classification. Extensive experiments demonstrate that CECF not only achieves superior performance but also serves as a flexible, plug-and-play enhancement for existing methods. We also provide empirical analyses, offering insights into when and how this high-dimensional causal modeling framework works for the edge classification.

cs.LG↗

arXiv2026

QiMeng-PRepair: Precise Code Repair via Edit-Aware Reward Optimization

Large Language Models (LLMs) achieve strong program repair performance but often suffer from over-editing, where excessive modifications overwrite correct code and hinder bug localization. We systematically quantify its impact and introduce precise repair task, which maximizes reuse of correct code while fixing only buggy parts. Building on this insight, we propose PRepair, a framework that mitigates over-editing and improves repair accuracy. PRepair has two components: Self-Breaking, which generates diverse buggy programs via controlled bug injection and min-max sampling, and Self-Repairing, which trains models with Edit-Aware Group Relative Policy Optimization (EA-GRPO) using an edit-aware reward to encourage minimal yet correct edits. Experiments show that PRepair improves repair precision by up to 31.4% under $\mathrm{fix}_1@1$, a metric that jointly considers repair correctness and extent, and significantly increases decoding throughput when combined with speculative editing, demonstrating its potential for precise and practical code repair.

cs.SE↗

arXiv2026

Efficient Reasoning with Balanced Thinking

Large Reasoning Models (LRMs) have shown remarkable reasoning capabilities, yet they often suffer from overthinking, expending redundant computational steps on simple problems, or underthinking, failing to explore sufficient reasoning paths despite inherent capabilities. These issues lead to inefficiencies and potential inaccuracies, limiting practical deployment in resource-constrained settings. Existing methods to mitigate overthinking, such as suppressing reflective keywords or adjusting reasoning length, may inadvertently induce underthinking, compromising accuracy. Therefore, we propose ReBalance, a training-free framework that achieves efficient reasoning with balanced thinking. ReBalance leverages confidence as a continuous indicator of reasoning dynamics, identifying overthinking through high confidence variance and underthinking via consistent overconfidence. By aggregating hidden states from a small-scale dataset into reasoning mode prototypes, we compute a steering vector to guide LRMs' reasoning trajectories. A dynamic control function modulates this vector's strength and direction based on real-time confidence, pruning redundancy during overthinking, and promoting exploration during underthinking. Extensive experiments conducted on four models ranging from 0.5B to 32B, and across nine benchmarks in math reasoning, general question answering, and coding tasks demonstrate that ReBalance effectively reduces output redundancy while improving accuracy, offering a general, training-free, and plug-and-play strategy for efficient and robust LRM deployment. Project page and code are available at https://rebalance-ai.github.io .

cs.AI↗

arXiv2026

OmniCellTOSG: The First Cell Text-Omic Signaling Graphs Dataset for Graph Language Foundation Modeling

With the rapid growth of large-scale single-cell omic datasets, omic foundation models (FMs) have emerged as powerful tools for advancing research in life sciences and precision medicine. However, most existing omic FMs rely primarily on numerical transcriptomic data by sorting genes as sequences, while lacking explicit integration of biomedical prior knowledge and signaling interactions that are critical for scientific discovery. Here, we introduce the Text-Omic Signaling Graph (TOSG), a novel data structure that unifies human-interpretable biomedical textual knowledge, quantitative omic data, and signaling network information. Using this framework, we construct OmniCellTOSG, a large-scale resource comprising approximately half million meta-cell TOSGs derived from around 80 million single-cell and single-nucleus RNA-seq profiles across organs and diseases. We further develop CellTOSG-FM, a multimodal graph language FM, to jointly analyze textual, omic and signaling network context. Across diverse downstream tasks, CellTOSG-FM outperforms existing omic FMs, and provides interpretable insights into disease-associated targets and signaling pathways.

cs.AI↗

arXiv2025

QiMeng-MuPa: Mutual-Supervised Learning for Sequential-to-Parallel Code Translation

The rise of GPU-based high-performance computing (HPC) has driven the widespread adoption of parallel programming models such as CUDA. Yet, the inherent complexity of parallel programming creates a demand for the automated sequential-to-parallel approaches. However, data scarcity poses a significant challenge for machine learning-based sequential-to-parallel code translation. Although recent back-translation methods show promise, they still fail to ensure functional equivalence in the translated code. In this paper, we propose \textbf{QiMeng-MuPa}, a novel \textbf{Mu}tual-Supervised Learning framework for Sequential-to-\textbf{Pa}rallel code translation, to address the functional equivalence issue. QiMeng-MuPa consists of two models, a Translator and a Tester. Through an iterative loop consisting of Co-verify and Co-evolve steps, the Translator and the Tester mutually generate data for each other and improve collectively. The Tester generates unit tests to verify and filter functionally equivalent translated code, thereby evolving the Translator, while the Translator generates translated code as augmented input to evolve the Tester. Experimental results demonstrate that QiMeng-MuPa significantly enhances the performance of the base models: when applied to Qwen2.5-Coder, it not only improves Pass@1 by up to 28.91% and boosts Tester performance by 68.90%, but also outperforms the previous state-of-the-art method CodeRosetta by 1.56 and 6.92 in BLEU and CodeBLEU scores, while achieving performance comparable to DeepSeek-R1 and GPT-4.1. Our code is available at https://github.com/kcxain/mupa.

cs.SE↗

arXiv2025

Deep Reinforcement Learning-Based Decision-Making Strategy Considering User Satisfaction Feedback in Demand Response Program

Demand response providers (DRPs) are intermediaries between the upper-level distribution system operator and the lower-level participants in demand response (DR) programs. Usually, DRPs act as leaders and determine electricity pricing strategies to maximize their economic revenue, while end-users adjust their power consumption following the pricing signals. However, this profit-seeking bi-level optimization model often neglects the satisfaction of end-users participating in DR programs. In addition, the detailed mathematical models underlying user decision-making strategy and satisfaction evaluation mechanism are typically unavailable to DRPs, posing significant challenges to conventional model-based solution methods. To address these issues, this paper designs a user-side satisfaction evaluation mechanism and proposes a multi-branch temporal fusion twin-delayed deep deterministic policy gradient (MBTF-TD3) reinforcement learning algorithm. User satisfaction feedback is incorporated into the reward function via a dynamically adjusted penalty term. The proposed MBTF structure effectively extracts temporal feature dependencies in the time-series observation data, and the dynamically adjusted penalty function successfully enhances the overall satisfaction level of users. Several experiments are conducted to validate the performance and the effectiveness of our proposed solution algorithm.

eess.SY↗

arXiv2025

Model-Agnostic Sentiment Distribution Stability Analysis for Robust LLM-Generated Texts Detection

The rapid advancement of large language models (LLMs) has resulted in increasingly sophisticated AI-generated content, posing significant challenges in distinguishing LLM-generated text from human-written language. Existing detection methods, primarily based on lexical heuristics or fine-tuned classifiers, often suffer from limited generalizability and are vulnerable to paraphrasing, adversarial perturbations, and cross-domain shifts. In this work, we propose SentiDetect, a model-agnostic framework for detecting LLM-generated text by analyzing the divergence in sentiment distribution stability. Our method is motivated by the empirical observation that LLM outputs tend to exhibit emotionally consistent patterns, whereas human-written texts display greater emotional variability. To capture this phenomenon, we define two complementary metrics: sentiment distribution consistency and sentiment distribution preservation, which quantify stability under sentiment-altering and semantic-preserving transformations. We evaluate SentiDetect on five diverse datasets and a range of advanced LLMs,including Gemini-1.5-Pro, Claude-3, GPT-4-0613, and LLaMa-3.3. Experimental results demonstrate its superiority over state-of-the-art baselines, with over 16% and 11% F1 score improvements on Gemini-1.5-Pro and GPT-4-0613, respectively. Moreover, SentiDetect also shows greater robustness to paraphrasing, adversarial attacks, and text length variations, outperforming existing detectors in challenging scenarios.

cs.CL↗

arXiv2025

Large Language Model Inference Acceleration: A Comprehensive Hardware Perspective

Large Language Models (LLMs) have demonstrated remarkable capabilities across various fields, from natural language understanding to text generation. Compared to non-generative LLMs like BERT and DeBERTa, generative LLMs like GPT series and Llama series are currently the main focus due to their superior algorithmic performance. The advancements in generative LLMs are closely intertwined with the development of hardware capabilities. Various hardware platforms exhibit distinct hardware characteristics, which can help improve LLM inference performance. Therefore, this paper comprehensively surveys efficient generative LLM inference on different hardware platforms. First, we provide an overview of the algorithm architecture of mainstream generative LLMs and delve into the inference process. Then, we summarize different optimization methods for different platforms such as CPU, GPU, FPGA, ASIC, and PIM/NDP, and provide inference results for generative LLMs. Furthermore, we perform a qualitative and quantitative comparison of inference performance with batch sizes 1 and 8 on different hardware platforms by considering hardware power consumption, absolute inference speed (tokens/s), and energy efficiency (tokens/J). We compare the performance of the same optimization methods across different hardware platforms, the performance across different hardware platforms, and the performance of different methods on the same hardware platform. This provides a systematic and comprehensive summary of existing inference acceleration work by integrating software optimization methods and hardware platforms. We point out that three trends (multimodality, inference-time compute, and higher inference energy efficiency) are promising to redefine the capabilities of edge artificial intelligence systems. Our project is available at https://dai.sjtu.edu.cn/project.html.

cs.AR↗