arXiv subjects

Chenning Li

Publications and source records attributed to Chenning Li.

12 recordsLinked to original sources

arXiv2026

ADR: An Agentic Detection System for Enterprise Agentic AI Security

We present the Agentic AI Detection and Response (ADR) system, the first large-scale, production-proven enterprise framework for securing AI agents operating through the Model Context Protocol (MCP). We identify three persistent challenges in this domain: (1) limited observability -- existing Endpoint Detection and Response (EDR) tools see file writes but not the agent reasoning, prompts, or causal chains linking intent to execution; (2) insufficient robustness -- static defenses constrained by pre-defined rules fail to generalize across diverse attack techniques and enterprise contexts; and (3) high detection costs -- LLM-based inference is prohibitively expensive at scale. ADR addresses these challenges via three components: the ADR Sensor for high-fidelity agentic telemetry, the ADR Explorer for systematic pre-deployment red teaming and hard-example generation, and the ADR Detector for scalable, two-tier online detection combining fast triage with context-aware reasoning. Deployed at Uber for over ten months, ADR has sustained reliable detection in production with growing adoption reaching over 7,200 unique hosts and processing over 10,000 agent sessions daily, uncovering hundreds of credential exposures across 26 categories and enabling a shift-left prevention layer (97.2% precision, 206 detected credentials). To validate the approach and enable community adoption, we introduce ADR-Bench (302 tasks, 17 techniques, 133 MCP servers), where ADR achieves zero false positives while detecting 67% of attacks -- outperforming three state-of-the-art baselines (ALRPHFS, GuardAgent, LlamaFirewall) by 2--4x in F1-score. On AgentDojo (public prompt injection benchmark), ADR detects all attacks with only three false alarms out of 93 tasks.

cs.AI↗

arXiv2026

AcademiClaw: When Students Set Challenges for AI Agents

Benchmarks within the OpenClaw ecosystem have thus far evaluated exclusively assistant-level tasks, leaving the academic-level capabilities of OpenClaw largely unexamined. We introduce AcademiClaw, a bilingual benchmark of 80 complex, long-horizon tasks sourced directly from university students' real academic workflows -- homework, research projects, competitions, and personal projects -- that they found current AI agents unable to solve effectively. Curated from 230 student-submitted candidates through rigorous expert review, the final task set spans 25+ professional domains, ranging from olympiad-level mathematics and linguistics problems to GPU-intensive reinforcement learning and full-stack system debugging, with 16 tasks requiring CUDA GPU execution. Each task executes in an isolated Docker sandbox and is scored on task completion by multi-dimensional rubrics combining six complementary techniques, with an independent five-category safety audit providing additional behavioral analysis. Experiments on six frontier models show that even the best achieves only a 55\% pass rate. Further analysis uncovers sharp capability boundaries across task domains, divergent behavioral strategies among models, and a disconnect between token consumption and output quality, providing fine-grained diagnostic signals beyond what aggregate metrics reveal. We hope that AcademiClaw and its open-sourced data and code can serve as a useful resource for the OpenClaw community, driving progress toward agents that are more capable and versatile across the full breadth of real-world academic demands. All data and code are available at https://github.com/GAIR-NLP/AcademiClaw.

cs.AI↗

arXiv2026

Prediction-Guided Control in Data Center Networks

In this paper, we design, implement, and evaluate Polyphony, a system to give network operators a new way to control and reduce the frequency of poor tail latency events in multi-class data center networks, on the time scale of minutes. Polyphony is designed to be complementary to other adaptive mechanisms like congestion control and traffic engineering, but targets different aspects of network operation that have previously been considered static. By contrast to Polyphony, prior model-free optimization methods work best when there are only a few relevant degrees of freedom and where workloads and measurements are stable, assumptions not present in modern data center networks. Polyphony develops novel methods for measuring, predicting, and controlling network quality of service metrics for a dynamically changing workload. First, we monitor and aggregate workloads on a network-wide basis; we use the result as input to an approximate counterfactual prediction engine that estimates the effect of potential network configuration changes on network quality of service; we apply the best candidate and repeat in a closed-loop manner aimed at rapidly and stably converging to a configuration that meets operator goals. Using CloudLab on a simple topology, we observe that Polyphony converges to tight SLOs within ten minutes, and re-stabilizes after large workload shifts within fifteen minutes, while the prior state of the art fails to adapt.

cs.NI↗

arXiv2025

Morph: ChirpTransformer-based Encoder-decoder Co-design for Reliable LoRa Communication

In this paper, we propose Morph, a LoRa encoder-decoder co-design to enhance communication reliability while improving its computation efficiency in extremely-low signal-to-noise ratio (SNR) situations. The standard LoRa encoder controls 6 Spreading Factors (SFs) to tradeoff SNR tolerance with data rate. SF-12 is the maximum SF providing the lowest SNR tolerance on commercial off-the-shelf (COTS) LoRa nodes. In Morph, we develop an SF-configuration based encoder to mimic the larger SFs beyond SF-12 while it is compatible with COTS LoRa nodes. Specifically, we manipulate four SF configurations of a Morph symbol to encode 2-bit data. Accordingly, we recognize the used SF configuration of the symbol for data decoding. We leverage a Deep Neural Network (DNN) decoder to fully capture multi-dimensional features among diverse SF configurations to maximize the SNR gain. Moreover, we customize the input size, neural network structure, and training method of the DNN decoder to improve its efficiency, reliability, and generalizability. We implement Morph with COTS LoRa nodes and a USRP N210, then evaluate its performance on indoor and campus-scale testbeds. Results show that we can reliably decode data at -28.8~dB SNR, which is 6.4~dB lower than the standard LoRa with SF-12 chirps. In addition, the computation efficiency of our DNN decoder is about 3x higher than state-of-the-art.

cs.NI↗

arXiv2025

KramaBench: A Benchmark for AI Systems on Data-to-Insight Pipelines over Data Lakes

Discovering insights from a real-world data lake potentially containing unclean, semi-structured, and unstructured data requires a variety of data processing tasks, ranging from extraction and cleaning to integration, analysis, and modeling. This process often also demands domain knowledge and project-specific insight. While AI models have shown remarkable results in reasoning and code generation, their abilities to design and execute complex pipelines that solve these data-lake-to-insight challenges remain unclear. We introduce KramaBench which consists of 104 manually curated and solved challenges spanning 1700 files, 24 data sources, and 6 domains. KramaBench focuses on testing the end-to-end capabilities of AI systems to solve challenges which require automated orchestration of different data tasks. KramaBench also features a comprehensive evaluation framework assessing the pipeline design and individual data task implementation abilities of AI systems. We evaluate 8 LLMs using our single-agent reference framework DS-Guru, alongside both open- and closed-source single- and multi-agent systems, and find that while current agentic systems may handle isolated data-science tasks and generate plausible draft pipelines, they struggle with producing working end-to-end pipelines. On KramaBench, the best system reaches only 55% end-to-end accuracy in the full data-lake setting. Even with perfect retrieval, the accuracy tops out at 62%. Leading LLMs can identify up to 42% of important data tasks but can only fully implement 20% of individual data tasks. Our code, reference framework, and data are available at https://github.com/mitdbg/KramaBench.

cs.DB↗

arXiv2025

Scalable Routing in a City-Scale Wi-Fi Network for Disaster Recovery

In this paper, we present a new city-scale decentralized mesh network system suited for disaster recovery and emergencies. When wide-area connectivity is unavailable or significantly degraded, our system, MapMesh, enables static access points and mobile devices equipped with Wi-Fi in a city to route packets via each other for intra-city connectivity and to/from any nodes that might have Internet access, e.g., via satellite. The chief contribution of our work is a new routing protocol that scales to millions of nodes, a significant improvement over prior work on wireless mesh and mobile ad hoc networks. Our approach uses detailed information about buildings from widely available maps--data that was unavailable at scale over a decade ago, but is widely available now--to compute paths in a scalable way.

cs.NI↗

arXiv2025

m4: A Learned Flow-level Network Simulator

Flow-level simulation is widely used to model large-scale data center networks due to its scalability. Unlike packet-level simulators that model individual packets, flow-level simulators abstract traffic as continuous flows with dynamically assigned transmission rates. While this abstraction enables orders-of-magnitude speedup, it is inaccurate by omitting critical packet-level effects such as queuing, congestion control, and retransmissions. We present m4, an accurate and scalable flow-level simulator that uses machine learning to learn the dynamics of the network of interest. At the core of m4 lies a novel ML architecture that decomposes state transition computations into distinct spatial and temporal components, each represented by a suitable neural network. To efficiently learn the underlying flow-level dynamics, m4 adds dense supervision signals by predicting intermediate network metrics such as remaining flow size and queue length during training. m4 achieves a speedup of up to 104$\times$ over packet-level simulation. Relative to a traditional flow-level simulation, m4 reduces per-flow estimation errors by 45.3% (mean) and 53.0% (p90). For closed-loop applications, m4 accurately predicts network throughput under various congestion control schemes and workloads.

cs.NI↗

arXiv2023

DriveTrack: A Benchmark for Long-Range Point Tracking in Real-World Videos

This paper presents DriveTrack, a new benchmark and data generation framework for long-range keypoint tracking in real-world videos. DriveTrack is motivated by the observation that the accuracy of state-of-the-art trackers depends strongly on visual attributes around the selected keypoints, such as texture and lighting. The problem is that these artifacts are especially pronounced in real-world videos, but these trackers are unable to train on such scenes due to a dearth of annotations. DriveTrack bridges this gap by building a framework to automatically annotate point tracks on autonomous driving datasets. We release a dataset consisting of 1 billion point tracks across 24 hours of video, which is seven orders of magnitude greater than prior real-world benchmarks and on par with the scale of synthetic benchmarks. DriveTrack unlocks new use cases for point tracking in real-world videos. First, we show that fine-tuning keypoint trackers on DriveTrack improves accuracy on real-world scenes by up to 7%. Second, we analyze the sensitivity of trackers to visual artifacts in real scenes and motivate the idea of running assistive keypoint selectors alongside trackers.

cs.CV↗

arXiv2023

Security Defense For Smart Contracts: A Comprehensive Survey

The blockchain technology has been used for recording state transitions of smart contracts - decentralized applications that can be invoked through external transactions. Smart contracts gained popularity and accrued hundreds of billions of dollars in market capitalization in recent years. Unfortunately, like all other programs, smart contracts are prone to security vulnerabilities that have incurred multimillion-dollar damages over the past decade. As a result, many automated threat mitigation solutions have been proposed to counter the security issues of smart contracts. These threat mitigation solutions include various tools and methods that are challenging to compare. This survey develops a comprehensive classification taxonomy of smart contract threat mitigation solutions within five orthogonal dimensions: defense modality, core method, targeted contracts, input-output data mapping, and threat model. We classify 133 existing threat mitigation solutions using our taxonomy and confirm that the proposed five dimensions allow us to concisely and accurately describe any smart contract threat mitigation solution. In addition to learning what the threat mitigation solutions do, we also show how these solutions work by synthesizing their actual designs into a set of uniform workflows corresponding to the eight existing defense core methods. We further create an integrated coverage map for the known smart contract vulnerabilities by the existing threat mitigation solutions. Finally, we perform the evidence-based evolutionary analysis, in which we identify trends and future perspectives of threat mitigation in smart contracts and pinpoint major weaknesses of the existing methodologies. For the convenience of smart contract security developers, auditors, users, and researchers, we deploy a regularly updated comprehensive open-source online registry of threat mitigation solutions.

cs.CR↗

arXiv2022

WiVelo: Fine-grained Walking Velocity Estimation for Wi-Fi Passive Tracking

Passive human tracking via Wi-Fi has been researched broadly in the past decade. Besides straight-forward anchor point localization, velocity is another vital sign adopted by the existing approaches to infer user trajectory. However, state-of-the-art Wi-Fi velocity estimation relies on Doppler-Frequency-Shift (DFS) which suffers from the inevitable signal noise incurring unbounded velocity errors, further degrading the tracking accuracy. In this paper, we present WiVelo\footnote{Code\&datasets are available at \textit{https://github.com/liecn/WiVelo\_SECON22}} that explores new spatial-temporal signal correlation features observed from different antennas to achieve accurate velocity estimation. First, we use subcarrier shift distribution (SSD) extracted from channel state information (CSI) to define two correlation features for direction and speed estimation, separately. Then, we design a mesh model calculated by the antennas' locations to enable a fine-grained velocity estimation with bounded direction error. Finally, with the continuously estimated velocity, we develop an end-to-end trajectory recovery algorithm to mitigate velocity outliers with the property of walking velocity continuity. We implement WiVelo on commodity Wi-Fi hardware and extensively evaluate its tracking accuracy in various environments. The experimental results show our median and 90\% tracking errors are 0.47~m and 1.06~m, which are half and a quarter of state-of-the-arts.

cs.HC↗

arXiv2022

NEC: Speaker Selective Cancellation via Neural Enhanced Ultrasound Shadowing

In this paper, we propose NEC (Neural Enhanced Cancellation), a defense mechanism, which prevents unauthorized microphones from capturing a target speaker's voice. Compared with the existing scrambling-based audio cancellation approaches, NEC can selectively remove a target speaker's voice from a mixed speech without causing interference to others. Specifically, for a target speaker, we design a Deep Neural Network (DNN) model to extract high-level speaker-specific but utterance-independent vocal features from his/her reference audios. When the microphone is recording, the DNN generates a shadow sound to cancel the target voice in real-time. Moreover, we modulate the audible shadow sound onto an ultrasound frequency, making it inaudible for humans. By leveraging the non-linearity of the microphone circuit, the microphone can accurately decode the shadow sound for target voice cancellation. We implement and evaluate NEC comprehensively with 8 smartphone microphones in different settings. The results show that NEC effectively mutes the target speaker at a microphone without interfering with other users' normal conversations.

cs.SD↗

arXiv2022

CurvingLoRa to Boost LoRa Network Capacity via Concurrent Transmission

LoRaWAN has emerged as an appealing technology to connect IoT devices but it functions without explicit coordination among transmitters, which can lead to many packet collisions as the network scales. State-of-the-art work proposes various approaches to deal with these collisions, but most functions only in high signal-to-interference ratio (SIR) conditions and thus does not scale to many scenarios where weak receptions are easily buried by stronger receptions from nearby transmitters. In this paper, we take a fresh look at LoRa's physical layer, revealing that its underlying linear chirp modulation fundamentally limits the capacity and scalability of concurrentLoRa transmissions. We show that by replacing linear chirps with their non-linear counterparts, we can boost the capacity of concurrent LoRa transmissions and empower the LoRa receiver to successfully receive weak transmissions in the presence of strong colliding signals. Such a non-linear chirp design further enables the receiver to demodulate fully aligned collision symbols - a case where none of the existing approaches can deal with. We implement these ideas in a holistic LoRaWAN stack based on the USRP N210 software-defined radio platform. Our head-to-head comparison with two state-of-the-art research systems and a standard LoRaWAN baseline demonstrates that CurvingLoRa improves the network throughput by 1.6-7.6x while simultaneously sacrificing neither power efficiency nor noise resilience. An open-source dataset and code will be made available before publication.

cs.NI↗