Description

本日紹介した論文の一覧

• Cheating Automatic LLM Benchmarks: Null Models Achieve High Win Rates
  http://arxiv.org/abs/2410.07137v1

• $\texttt{ModSCAN}$: Measuring Stereotypical Bias in Large
  Vision-Language Models from Vision and Language Modalities
  http://arxiv.org/abs/2410.06967v1

• Privately Counting Partially Ordered Data
  http://arxiv.org/abs/2410.06881v1

• On Wagner's k-Tree Algorithm Over Integers
  http://arxiv.org/abs/2410.06856v1

• On the Security and Design of Cryptosystems Using Gabidulin-Kronecker
  Product Codes
  http://arxiv.org/abs/2410.06849v1

• Root Defence Strategies: Ensuring Safety of LLM at the Decoding Level
  http://arxiv.org/abs/2410.06809v1

• Diffuse or Confuse: A Diffusion Deepfake Speech Dataset
  http://arxiv.org/abs/2410.06796v1

• Mind Your Questions Towards Backdoor Attacks on Text-to-Visualization
  Models
  http://arxiv.org/abs/2410.06782v1

• MERGE: Matching Electronic Results with Genuine Evidence for verifiable
  voting in person at remote locations
  http://arxiv.org/abs/2410.06705v1

• How hard can it be? Quantifying MITRE attack campaigns with attack trees
  and cATM logic
  http://arxiv.org/abs/2410.06692v1

• Bots can Snoop: Uncovering and Mitigating Privacy Risks of Bots in Group
  Chats
  http://arxiv.org/abs/2410.06587v1

• Can DeepFake Speech be Reliably Detected?
  http://arxiv.org/abs/2410.06572v1

• Signal Watermark on Large Language Models
  http://arxiv.org/abs/2410.06545v1

• Gumbel Rao Monte Carlo based Bi-Modal Neural Architecture Search for
  Audio-Visual Deepfake Detection
  http://arxiv.org/abs/2410.06543v1

• On the Security of Bitstream-level JPEG Encryption with Restart Markers
  http://arxiv.org/abs/2410.06522v1

• MORSE: An Efficient Homomorphic Secret Sharing Scheme Enabling
  Non-Linear Operation
  http://arxiv.org/abs/2410.06514v1

• WAPITI: A Watermark for Finetuned Open-Source LLMs
  http://arxiv.org/abs/2410.06467v1

• Hallucinating AI Hijacking Attack: Large Language Models and Malicious
  Code Recommenders
  http://arxiv.org/abs/2410.06462v1

• Multi-label Classification for Android Malware Based on Active Learning
  http://arxiv.org/abs/2410.06444v1

なお、ポッドキャスト内で紹介する内容は、各論文の概要を日本語で解説したもので、論文概要の著作権は論文著者に帰属します。