proomt

Search

Search posts, papers, and topics

All posts

Hugging Face Daily PapersHuiyuan Liu, Zhiming Ma, Yanxing Liu1 min readpaperadvanced

TeleAntiFraud 2.0: A Refreshable, Profile-Grounded, and Audio-Based Benchmark for Telecom Fraud Detection

Read on arxiv.org

Summary

The paper introduces TeleAntiFraud 2.0, a monthly refreshed, frozen audio benchmark of 900 Chinese telecom calls (600 fraud, 300 near‑domain non‑fraud) built with a Mixed‑Tree generation pipeline. Experiments show models that score perfectly on unrelated negatives fall to ~0.66 Macro‑F1 on near‑domain cases, exposing shortcut learning and prediction collapse.

  • TeleAntiFraud 2.0 provides a monthly refreshed, frozen audio benchmark of 900 Chinese calls (600 fraud, 300 near‑domain non‑fraud) with full provenance.
  • The Mixed‑Tree Anti‑Fraud Generation Pipeline creates profile‑grounded scenarios and expands them via mixed‑tree generation for realistic dialogues.
  • Classifiers that achieve perfect Macro‑F1 on unrelated negatives drop to ~0.66 on near‑domain negatives, revealing reliance on topic cues.
  • ASR+LLM pipelines exhibit class‑prior shortcuts and prediction collapse, underscoring the need for collapse‑aware reporting.

Engineers building audio‑based telecom fraud detectors need realistic, refreshable benchmarks with near‑domain negatives to avoid misleading performance claims.

8/10
audionlpspeech recognitionfraud detectiondatasets

Related reading

  1. FRAUDSkill: Structured Frozen-Weight Skill Optimization for Audio Anti-Fraud Detection

    FRAUDSkill is a framework that keeps a pretrained audio‑language model frozen and learns an external layer of skill programs, routing policies, and decision rules to meet a structured anti‑fraud detection protocol. On the TeleAntiFraud benchmark it reaches 73.5% Macro‑F1 (≈32% improvement) while cutting invalid predictions to 1.94%.

    Hugging Face Daily Papersarxiv.org1 minpaper

  2. webaudio fingerprinting on alibaba

    Firefox made WebAudio output deterministic in v118, collapsing most variance to three CPU‑dependent buckets (x86 without FMA, x86/x64 with FMA, ARM NEON). Telemetry shows 99.24% of users fall into these buckets; a long tail of 23 distinct values affects <0.1% of users. Ongoing bugs aim to merge the remaining buckets. Alibaba’s fingerprint script was extracted (two SHA‑256 hashes) and shown to be…

    Mozilla Automation Teamritter.vg3 min

  3. E2A-Bench: Benchmarking Evidence-to-Action Reliability in Financial Chart Reasoning

    E2A‑Bench is a 969‑query benchmark for financial chart reasoning that evaluates vision‑language models across a full evidence‑to‑action chain using four metrics (UCR, RCI, ECI, NDR). Experiments on 20 VLMs expose hidden failures: low‑UCR models have only 6.4 % directional coverage, oracle‑aided verification cuts unsupported claims but can kill coverage, and fine‑tuning inflates BUY:SELL ratios by…

    Hugging Face Daily Papersarxiv.org1 minpaper

  5. Article: Architecting Secure and Scalable Facial Verification Systems

    A real‑world post‑mortem of a high‑volume face verification service that moved from a naïve synchronous API to an async, layered pipeline (edge validation, preprocessing, decoupled detection/verification, decision engine) to achieve 8.5k rpm, p99 < 1.8 s, 30 % cost savings, and strict privacy controls.

    InfoQinfoq.com15 min

  6. StepAudio 3 Realtime Technical Report

    StepAudio 3 Realtime is an audio‑language foundation model that runs a continuous listen‑converse‑think‑act loop. It introduces Deep Perception for rich acoustic cue extraction, Seamless Duplex for handling pauses/back‑channels, and a Think‑While‑Speaking mechanism that lets the model reason in parallel with speech output. On benchmarks it scores 73.0 macro avg on StepAudioChat, 90.6 on MMSU, 98.…

    Hugging Face Daily Papersarxiv.org2 minpaper