Paper: arXiv 2407.14573
Authors: Orson Mengara
Abstract
Since the advent of generative artificial intelligence, every company and researcher has been rushing to develop their own generative models, whether commercial or not. Given the large number of users of these powerful new tools, there is currently no intrinsically verifiable way to explain from the ground up what happens when LLMs (large language models) learn. For example, those based on automatic speech recognition systems, which have to rely on huge and astronomical amounts of data collected from all over the web to produce fast and efficient results, In this article, we develop a backdoor attack called MarketBackFinal 2.0, based on acoustic data poisoning, MarketBackFinal 2.0 is mainly based on modern stock market models. In order to show the possible vulnerabilities of speech-based transformers that may rely on LLMs.
Complexity vs Empirical Score
- Math Complexity: 8.5/10
- Empirical Rigor: 2.0/10
- Quadrant: Lab Rats — high math complexity, limited empirical testing
Why this score: The paper introduces advanced financial mathematics (e.g., stochastic calculus, optimal transport, Bayesian optimization) but lacks any backtesting data, performance metrics, or implementation details, focusing instead on theoretical attack methodology.
Research Flowchart
flowchart TD A["Research Goal:<br>Vulnerability of Speech LLMs"] --> B["Data: <br>Acoustic Datasets & Stock Market Data"] B --> C["Methodology:<br>Bayesian Optimization &<br>Acoustic Poisoning"] C --> D["Computation:<br>Backdoor Injection<br>MarketBackFinal 2.0"] D --> E["Outcome:<br>Identified Vulnerabilities<br>in Speech Transformers"]
Evidence Card
What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card
Proposes
The paper proposes a backdoor attack called MarketBackFinal 2.0, based on acoustic data poisoning and modern stock market models.
Source sentence
For example, those based on automatic speech recognition systems, which have to rely on huge and astronomical amounts of data collected from all over the web to produce fast and efficient results, In this article, we develop a backdoor attack called MarketBackFinal 2.0, based on acoustic data poisoning, MarketBackFinal 2.0 is mainly based on modern stock market models.
Assumes 3
- The model owner is the defender and the trusted third party is the attacker in the threat model. model
Source sentence
We assume that the model owner and the trusted third party play the roles of defender and attacker, respectively, in the threat model.
- In a white‑box setting the adversary can access and modify the dataset, model parameters and structure. other
Source sentence
In a white-box environment, the adversary understands and controls the target dataset and model, including the ability to access and modify the dataset, as well as parameters and the structure of the model.
- In a black‑box setting the adversary can only manipulate part of the training data and has no knowledge of model structure or parameters. other
Source sentence
However, in the stricter framework of the black-box, the attacker is only able to manipulate part of the training data but has no knowledge of the structure and parameters of the target model, such as weights, hyperparameters, configurations, etc.
Data
- ESC-50 environmental dataset benchmark public
Source sentence
We use the ESC-50 environmental dataset [66], the dataset ESC-50 (environmental), is a labeled collection of 2,000 environmental sounds, which comprises five-second recordings sorted into 50 semantic categories (with 40 examples per category) that are further divided into five main groups: animals; natural soundscapes and water sounds; humans, non-vocal sounds; indoor/domestic sounds; and outdoor/urban sounds.
Costs
Not applicable
Finds 2
- The backdoor attack achieves near‑perfect attack success rate (100%) while maintaining high benign accuracy (≈94–99%) across multiple pre‑trained speech models. performance figure from elsewhere in the paper
Source sentence
Hugging Face Models Benign Accuracy (BA) Attack Success Rate (ASR) wav2vec 2.0 94.73% 100% whisper (OpenAI) 95.03% 100% HuBERT 95.21% 100% facebook/w2v-bert-2.0(Facebook) 98.96% 100% facebook/mms-1b-all 93.31% 100% Speech Encoder Decoder 96.12% 100% Data2vec 99.12% 100% 2 ESC-50 environmental dataset.
- The authors state that MarketBackFinal 2.0 easily misleads almost all tested Hugging Face transformer models. other figure from elsewhere in the paper
Source sentence
We can see that our backdoor 18Hugging Face Transformers THE FINAL PREPRINT , JOURNAL NAME ——-, VOL. .., NO. ..., MONTH 2024 5 attack easily manages to mislead these models (readers are invited to test 19), other Hugging Face models; as far as we know, we’ve managed to fool almost all these models.
Checked by others
No other paper in the archive has been found to support or contradict these findings (compared with 6 related papers). Coverage is partial, so this is neither replication nor refutation.
Code
No repository by the authors found in the paper.
Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.
Cited References (125)
- Lifting the Heston model (2019)
- On simulation of rough Volterra stochastic volatility models (2021) — link
- Volatility models in practice: Rough, Path-dependent or Markovian? — → 2401.03345
- Volatility is rough (2018)
- Actuarial applications of natural language processing using transformers: Case studies for using text features in an actuarial context (2024)
- A partial rough path space for rough volatility (2024)
- Rough volatility: fact or artefact? (2024)
- Multiscaling and rough volatility: An empirical investigation (2022)
and 117 more
- Multifactor approximation of rough volatility models (2019)
- Short-term at-the-money asymptotics under stochastic volatility models (2019)
- The characteristic function of rough Heston models (2019)
- The microstructural foundations of leverage effect and rough volatility (2018)
- Perfect hedging in rough Heston models (2018)
- Markovian structure of the Volterra Heston model (2019)
- Explainable bayesian optimization (2024) — link
- The unreasonable effectiveness of optimal transport in economics (2021) — link
- Functional central limit theorems for rough volatility (2024)
- Rough volatility (2023)
- New and robust drift approximations for the LIBOR market model (2008)
- How to keep text private? A systematic review of deep learning methods for privacy-preserving natural language processing (2023)
- Backdoor attacks and defenses in federated learning: Survey, challenges and future research directions (2024)
- A survey on XAI for beyond 5G security: technical aspects, use cases, challenges and research directions (2022) — link
- A survey on llm-gernerated text detection: Necessity, methods, and future directions (2023) — link
- Machine and deep learning for iot security and privacy: applications, challenges, and future directions (2022)
- Privacy-preserving machine learning: Methods, challenges and directions (2021) — link
- Gpt (generative pre-trained transformer)–a comprehensive review on enabling technologies, potential applications, emerging challenges, and future directions (2024)
- A survey of privacy attacks in machine learning (2023)
- Large language models for software engineering: A systematic literature review (2023) — link
- A survey on large language model (llm) security and privacy: The good, the bad, and the ugly (2024)
- Blockchain-based federated learning: A comprehensive survey (2021) — link
- A survey of safety and trustworthiness of large language models through the lens of verification and validation (2024)
- Sleeper agents: Training deceptive llms that persist through safety training (2024) — link
- From chatgpt to threatgpt: Impact of generative ai in cybersecurity and privacy (2023)
- A review of speech-centric trustworthy machine learning: Privacy, safety, and fairness (2023)
- Trustllm: Trustworthiness in large language models (2024) — link
- On the opportunities and risks of foundation models (2021) — link
- Deep hedging: learning to simulate equity option markets (2019) — link
- Ai in finance: challenges, techniques, and opportunities (2022)
- Sustainable ai: Environmental implications, challenges and opportunities (2022)
- Large language models and knowledge graphs: Opportunities and challenges (2023) — link
- AI and 6G into the metaverse: Fundamentals, challenges and future research trends (2024)
- Optimal static-dynamic hedges for exotic options under convex risk measures (2009)
- A survey on optimal transport for machine learning: Theory and applications (2021) — link
- Bayesian optimization of ESG financial investments — → 2303.01485
- A tutorial on Bayesian optimization (2018) — link
- The Art of Deception: Robust Backdoor Attack using Dynamic Stacking of Triggers (2024) — link
- The last Dance: Robust backdoor attack via diffusion models and bayesian approach (2024) — link
- Trading Devil: Robust backdoor attack via Stochastic investment models and Bayesian approach — → 2406.10719
- ESC: Dataset for environmental sound classification (2015)
- Robust speech recognition via large-scale weak supervision (2023)
- Seamless: Multilingual Expressive and Streaming Speech Translation (2023) — link
- Scaling speech technology to 1,000+ languages (2024)
- wav2vec 2.0: A framework for self-supervised learning of speech representations (2020)
- Hubert: Self-supervised speech representation learning by masked prediction of hidden units (2021)
- Data2vec: A general framework for self-supervised learning in speech, vision and language (2022)
- Wav2seq: Pre-training speech-to-text encoder-decoder models using pseudo languages (2023)
- Theory and practice of model risk management (2002)
- Certified Robustness to Data Poisoning in Gradient-Based Training (2024) — link
- Generalization Bound and New Algorithm for Clean-Label Backdoor Attack (2024) — link
- A comprehensive overview of backdoor attacks in large language models within communication networks (2024)
- Can you hear it? backdoor attacks via ultrasonic triggers (2022)
- Audio-domain position-independent backdoor attack via unnoticeable triggers (2022)
- Bayesian inference based on stationary fokker-planck sampling (2010)
- A comprehensive survey on applications of transformers for deep learning tasks (2023)
- Hugging face (2022)
- Opportunistic Backdoor Attacks: Exploring Human-imperceptible Vulnerabilities on Speech Recognition Systems (2022)
- Badgpt: Exploring security vulnerabilities of chatgpt via backdoor attacks to instructgpt (2023) — link
- Badpre: Task-agnostic backdoor attacks to pre-trained nlp foundation models (2021) — link
- Hidden killer: Invisible textual backdoor attacks with syntactic trigger (2021) — link
- Attacks in adversarial machine learning: A systematic survey from the life-cycle perspective (2023) — link
- A Survey of Backdoor Attacks and Defenses on Large Language Models: Implications for Security Measures (2024) — link
- Large language models in finance: A survey — → 2311.10723
- A Survey of Large Language Models for Financial Applications: Progress, Prospects and Challenges — → 2406.11903
- A survey of large language models in finance (finllms) (2024) — link
- Security and privacy challenges of large language models: A survey (2024) — link
- A survey on deep learning for financial risk prediction (2021)
- Delta hedging in financial engineering: Towards a model-free approach (2010)
- Preliminary remarks on option pricing and dynamic hedging (2012)
- Taking the human out of the loop: A review of Bayesian optimization (2015)
- Video Diffusion Models: A Survey (2024) — link
- A survey on video diffusion models (2023) — link
- Text-to-image diffusion models in generative ai: A survey (2023) — link
- A survey on generative diffusion models (2024)
- Diffusion models: A comprehensive survey of methods and applications (2023)
- Girsanov’s Theorem (2016)
- Implicit Diffusion: Efficient Optimization through Stochastic Sampling (2024) — link
- Large language models are efficient learners of noise-robust speech recognition (2024) — link
- Detectors for safe and reliable llms: Implementations, uses, and limitations (2024) — link
- Universal vulnerabilities in large language models: Backdoor attacks for in-context learning (2024) — link
- Foundational challenges in assuring alignment and safety of large language models (2024) — link
- Training compute-optimal large language models (2022) — link
- The Shadow of Fraud: The Emerging Danger of AI-powered Social Engineering and its Possible Cure (2024) — link
- Comparison of Static Application Security Testing Tools and Large Language Models for Repo-level Vulnerability Detection (2024) — link
- Imposter. AI: Adversarial Attacks with Hidden Intentions towards Aligned Large Language Models (2024) — link
- Targeted Latent Adversarial Training Improves Robustness to Persistent Harmful Behaviors in LLMs (2024) — link
- Operationalizing a Threat Model for Red-Teaming Large Language Models (LLMs) (2024) — link
- Deep reinforcement learning for automated stock trading: An ensemble strategy (2020)
- Financial statement analysis with large language models — → 2407.17866
- BaThe: Defense against the Jailbreak Attack in Multimodal Large Language Models by Treating Harmful Instruction as Backdoor Trigger (2024) — link
- Transferring Backdoors between Large Language Models by Knowledge Distillation (2024) — link
- BadMerging: Backdoor Attacks Against Model Merging (2024) — link
- A rough path perspective on renormalization (2019)
- Learning tensor networks with parameter dependence for Fourier-based option pricing — → 2405.00701
- Fast Fourier transform option pricing with stochastic interest rate, stochastic volatility and double jumps (2013)
- An FFT-network for L 'e vy option pricing (2011)
- Pricing Israeli Option with Time-changed Compensation by an FFT-Based High-order Multinomial Tree in L 'e vy Markets (2022)
- A Fourier transform method for spread option pricing (2010)
- Signature volatility models: pricing and hedging with Fourier (2024)
- Ruin probabilities in the Cram 'e r–Lundberg model with temporarily negative capital (2020)
- The Cram 'e r-Lundberg model with a fluctuating number of clients (2023)
- Ruin probability in the Cram 'e r–Lundberg model with risky investments (2011)
- Trading and arbitrage in cryptocurrency markets (2020)
- Critical probabilistic characteristics of the Cram 'e r model for primes and arithmetical properties (2024)
- Neural McKean-Vlasov Processes: Distributional Dependence in Diffusion Processes (2024)
- Pathwise McKean-Vlasov Theory with Additive Noise (2018) — link
- Control of McKean–Vlasov dynamics versus mean field games (2013)
- Learning high-dimensional McKean–Vlasov forward-backward stochastic differential equations with general distribution dependence (2024)
- McKean–Vlasov type stochastic differential equations arising from the random vortex method (2022)
- Large deviations and exit-times for reflected McKean–Vlasov equations with self-stabilising terms and superlinear drifts (2022)
- Maximum principle for optimal control of McKean-Vlasov FBSDEs with L 'e vy process via the differentiability with respect to probability law (2019)
- The rough Hawkes Heston stochastic volatility model (2024)
- Volatility options in rough volatility models (2020)
- On spatially irregular ordinary differential equations and a pathwise volatility modelling framework (2019) — link
- Affine rough models (2018) — link
- A geometric perspective on diffusion models (2023) — link
7 of 125 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.