arXiv subjects

Jiaxin Peng

Publications and source records attributed to Jiaxin Peng.

9 recordsLinked to original sources

arXiv2026

World Engine: Towards the Era of Post-Training for Autonomous Driving

Autonomous vehicles must operate safely in the real world, where errors can have severe consequences. Although modern end-to-end driving policies excel in routine scenarios, their reliability is limited by the scarcity of safety-critical ``long-tail'' events in real driving datasets. These rare interactions define the practical safety boundary of the learned policy, yet they are difficult to collect at scale in the real world. Here we show that this fundamental limitation can be addressed by post-training pre-trained driving models on synthesized high-stakes interactions. We introduce World Engine, a generative framework that reconstructs high-fidelity interactive environments from real-world logs and systematically extrapolates them into realistic safety-critical variations. This paradigm enables reinforcement-based post-training to align policies with safety constraints, circumventing the physical risks inherent in real-world exploration. On a public benchmark built on nuPlan, World Engine substantially reduces failures in rare safety-critical scenarios and yields significantly larger gains than scaling pre-training data alone. Furthermore, when deployed on a production-scale autonomous driving system, the resulting policy reduces simulated collisions and demonstrates measurable improvements in on-road testing, showing that post-training on synthesized, safety-critical interactions offers a scalable and effective pathway to safer autonomous driving. The full codebase suite, including training, is released to the public.

cs.RO↗

arXiv2026

Lingering Sweetness of Ethanol Clusters: Sensory Discovery and Objective Discrimination by Impedance-Based Electronic Tongue

Ethanol is conventionally perceived only as a pungent tastant, while the potential sweet properties of ethanol clusters have remained unrecognized. Here we show that ethanol tetramers exhibit a unique time dependent lingering sweetness, distinct from the immediate upupfront sweetness of conventional sweeteners. To objectively verify this phenomenon, we developed an impedance based bionic electronic tongue that monitors changes at a lipid polymer membrane interface. Sensory evaluation revealed increasing ethanol cluster content from 4.83% to 29.53% prolonged sweetness duration from 10s to 25s, whereas sweetness intensity remained low to moderate (1.0-1.7 on a 10 point scale). Using this electronic tongue, we achieved objective discrimination between upfront and lingering sweetness,full band phase angle scanning identified optimal frequencies of 0.933Hz (t=0.171s) for xylitol (fast binding and dissociation) and 0.03Hz (t=5.305s) for ethanol clusters (slow binding and dissociation). Phase angle responses correlated strongly with sensory sweetness duration. In natural aged Chinese Baijiu (1-20 years), low frequency phase angle signals positively correlated with sensory lingering sweetness scores, confirming that ethanol cluster accumulation during aging underpins the sweetness of Baijiu. This work expands the classic AH-B sweetness theory from single molecules to supramolecular clusters and provides an analytical platform for probing the temporal dimension of flavor perception.

physics.atom-ph↗

arXiv2026

Pairwise Preference Reward and Group-Based Diversity Enhancement for Superior Open-Ended Generation

Current reinforcement learning(RL) methods are broadly applicable and powerful in verifiable settings where scalar rewards can be provided. However, in open-ended generation tasks, verifying the correctness of responses remains challenging, and training reward models incurs substantial computational and annotation costs. Moreover, reinforcement learning (RLVR) often leads to diversity collapse and produces stereotypical or rigid outputs, outcomes that are particularly undesirable in open-domain scenarios. We propose Pairwise Preference Reward and Group-based Diversity Enhancement (PPR-GDE), a RL method that is more suitable for open-ended generation. PPR-GDE does not require scalar rewards and incorporates group-level diversity into the reward signal, it preserves the comparative structure of subjective evaluation through a pairwise preference reward, mitigates judge position bias via repeated comparisons with swapped response order, and introduces a group-based diversity reward that explicitly encourages semantic dispersion within a response group, all of these reward signals are integrated into a unified group-relative policy optimization objective. We instantiate PPR-GDE on role-playing task, experiments show that PPR-GDE achieves a better alignment quality as well as expressive diversity than strong RL baselines. Further analysis shows that pairwise preference is critical for preference alignment in subjective perspective, while the diversity metric plays an essential role in achieving superior expressive diversity and broader semantic coverage.

cs.AI↗

arXiv2025

T2R-bench: A Benchmark for Generating Article-Level Reports from Real World Industrial Tables

Extensive research has been conducted to explore the capabilities of large language models (LLMs) in table reasoning. However, the essential task of transforming tables information into reports remains a significant challenge for industrial applications. This task is plagued by two critical issues: 1) the complexity and diversity of tables lead to suboptimal reasoning outcomes; and 2) existing table benchmarks lack the capacity to adequately assess the practical application of this task. To fill this gap, we propose the table-to-report task and construct a bilingual benchmark named T2R-bench, where the key information flow from the tables to the reports for this task. The benchmark comprises 457 industrial tables, all derived from real-world scenarios and encompassing 19 industry domains as well as 4 types of industrial tables. Furthermore, we propose an evaluation criteria to fairly measure the quality of report generation. The experiments on 25 widely-used LLMs reveal that even state-of-the-art models like Deepseek-R1 only achieves performance with 62.71 overall score, indicating that LLMs still have room for improvement on T2R-bench.

cs.CL↗

arXiv2025

Technical Report of TeleChat2, TeleChat2.5 and T1

We introduce the latest series of TeleChat models: \textbf{TeleChat2}, \textbf{TeleChat2.5}, and \textbf{T1}, offering a significant upgrade over their predecessor, TeleChat. Despite minimal changes to the model architecture, the new series achieves substantial performance gains through enhanced training strategies in both pre-training and post-training stages. The series begins with \textbf{TeleChat2}, which undergoes pretraining on 10 trillion high-quality and diverse tokens. This is followed by Supervised Fine-Tuning (SFT) and Direct Preference Optimization (DPO) to further enhance its capabilities. \textbf{TeleChat2.5} and \textbf{T1} expand the pipeline by incorporating a continual pretraining phase with domain-specific datasets, combined with reinforcement learning (RL) to improve performance in code generation and mathematical reasoning tasks. The \textbf{T1} variant is designed for complex reasoning, supporting long Chain-of-Thought (CoT) reasoning and demonstrating substantial improvements in mathematics and coding. In contrast, \textbf{TeleChat2.5} prioritizes speed, delivering rapid inference. Both flagship models of \textbf{T1} and \textbf{TeleChat2.5} are dense Transformer-based architectures with 115B parameters, showcasing significant advancements in reasoning and general task performance compared to the original TeleChat. Notably, \textbf{T1-115B} outperform proprietary models such as OpenAI's o1-mini and GPT-4o. We publicly release \textbf{TeleChat2}, \textbf{TeleChat2.5} and \textbf{T1}, including post-trained versions with 35B and 115B parameters, to empower developers and researchers with state-of-the-art language models tailored for diverse applications.

cs.CL↗

arXiv2025

QoS-Efficient Serving of Multiple Mixture-of-Expert LLMs Using Partial Runtime Reconfiguration

The deployment of mixture-of-experts (MoE) large language models (LLMs) presents significant challenges due to their high memory demands. These challenges become even more pronounced in multi-tenant environments, where shared resources must accommodate multiple models, limiting the effectiveness of conventional virtualization techniques. This paper addresses the problem of efficiently serving multiple fine-tuned MoE-LLMs on a single-GPU. We propose a serving system that employs \textit{similarity-based expert consolidation} to reduce the overall memory footprint by sharing similar experts across models. To ensure output quality, we introduce \textit{runtime partial reconfiguration}, dynamically replacing non-expert layers when processing requests from different models. As a result, our approach achieves a competitive output quality while maintaining throughput comparable to serving a single model while incurring a negligible increase in time-to-first-token (TTFT). Experiments on a server with a single NVIDIA A100 GPU (80GB) using Mixtral-8x7B models demonstrate an 85\% average reduction in turnaround time compared to NVIDIA's multi-instance GPU (MIG). Furthermore, experiments on Google's Switch Transformer Base-8 model with up to four variants demonstrate the scalability and resilience of our approach in maintaining output quality compared to other model merging baselines, highlighting its effectiveness.

cs.LG↗

arXiv2024

TeleChat Technical Report

In this technical report, we present TeleChat, a collection of large language models (LLMs) with parameters of 3 billion, 7 billion and 12 billion. It includes pretrained language models as well as fine-tuned chat models that is aligned with human preferences. TeleChat is initially pretrained on an extensive corpus containing a diverse collection of texts from both English and Chinese languages, including trillions of tokens. Subsequently, the model undergoes fine-tuning to align with human preferences, following a detailed methodology that we describe. We evaluate the performance of TeleChat on various tasks, including language understanding, mathematics, reasoning, code generation, and knowledge-based question answering. Our findings indicate that TeleChat achieves comparable performance to other open-source models of similar size across a wide range of public benchmarks. To support future research and applications utilizing LLMs, we release the fine-tuned model checkpoints of TeleChat's 7B and 12B variant, along with code and a portion of our pretraining data, to the public community.

cs.CL↗

arXiv2017

Integrated Nanophotonics Architecture for Residue Number System Arithmetic

Residue number system (RNS) enables dimensionality reduction of an arithmetic problem by representing a large number as a set of smaller integers, where the number is decomposed by prime number factorization using the moduli as basic functions. These reduced problem sets can then be processed independently and in parallel, thus improving computational efficiency and speed. Here we show an optical RNS hardware representation based on integrated nanophotonics. The digit-wise shifting in RNS arithmetic is expressed as spatial routing of an optical signal in 2x2 hybrid photonic-plasmonic switches. Here the residue is represented by spatially shifting the input waveguides relative to the routers outputs, where the moduli are represented by the number of waveguides. By cascading the photonic 2x2 switches, we design a photonic RNS adder and a multiplier forming an all-to-all sparse directional network. The advantage of this photonic arithmetic processor is the short (10's ps) computational execution time given by the optical propagation delay through the integrated nanophotonic router. Furthermore, we show how photonic processing in-the-network leverages the natural parallelism of optics such as wavelength-division-multiplexing or optical angular momentum in this RNS processor. A key application for photonic RNS is the functional analysis convolution with widespread usage in numerical linear algebra, computer vision, language- image- and signal processing, and neural networks.

cs.ET↗

arXiv2017

MODetector (MOD): A Dual-Functional Optical Modulator-Detector for On-Chip Communication

Physical challenges at the device and interconnect level limit both network and computing energy efficiency. While photonics is being considered to address interconnect bottlenecks, optical routing is still limited by electronic circuitry, requiring substantial overhead for optical-electrical-optical conversion. Here we show a novel design of an integrated broadband photonic-plasmonic hybrid device termed MODetector featuring dual light modulation and detection function to act as an optical transceiver in the photonic network-on-chip. With over 10 dB extinction ratio and 0.8 dB insertion loss at the modulation state, this MODetector provides 0.7 W/A responsivity in the detection state with 36 ps response time. This multi-functional device: (i) eliminates OEO conversion, (ii) reduces optical losses from photodetectors when not needed, and (iii) enables cognitive routing strategies for network-on-chips.

physics.app-ph↗