ReCast: Contract-Preserving Protection for Fixed-Interface Multimodal Reasoning
cs.CR, cs.AI, cs.CL
Submitted: 2026-10-01
Updated: 2026-10-01
Code: https://github.com/deepseek-ai/deepseek-harness
Terminology
Sources
- Qwen3-VL Technical Report
- Privacy Preserving Visual Question Answering
- Safeguarding Privacy in Edge Speech Understanding with Tiny Foundation Models
- DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence
- QLoRA: Efficient Finetuning of Quantized LLMs
- Can LLMs get help from other LLMs without revealing private information?
- Qwen3-TTS Technical Report
- Privacy-R1: Privacy-Aware Multi-LLM Agent Collaboration via Reinforcement Learning
- Leveraging Semantic Triples for Private Document Generation with Local Differential Privacy Guarantees
- Qwen3-ASR Technical Report
- Beyond Direct Identifiers: Probabilistic Privacy Risk Estimation for Privacy-Conscious LLM Query Delegation
- Synthesize Privacy-Preserving High-Resolution Images via Private Textual Intermediaries
- Towards Spoken Mathematical Reasoning: Benchmarking Speech-based Models over Multi-faceted Math Problems
- ChartX & ChartVLM: A Versatile Benchmark and Foundation Model for Complicated Chart Reasoning
- Qwen2.5-Omni Technical Report
- Qwen3 Technical Report
- CoGenesis: A Framework Collaborating Large and Small Language Models for Secure Context-Aware Instruction Following
Related papers
- SoK: AI-Augmented Binary Reversing
- Relaxed Sender Anonymity for CBDC Interbank Settlement: A Zero-Knowledge Approach on Permissioned EVM
- Calibration-Family Overfit: Why Trusted Sabotage Monitors Don't Transfer Across Lineages
- Efficient Fuzzy PSI under One-Sided Assumptions
- Sealing the Audit-Runtime Gap for LLM Skills
- Token Composition: A Graph Based on EVM Logs