TRACER: Token ReAssignment for Concept ERasure in Generative Recommendation
cs.IR, cs.AI, cs.CL, cs.LG
Submitted: 2026-06-05
Updated: 2026-11-02
Terminology
Sources
- CURE:Circuit-Aware Unlearning for LLM-based Recommendation
- MU-Bench: A Multitask Multimodal Benchmark for Machine Unlearning
- A Mechanistic Perspective and Circuit-Guided Difficulty Metric for Unlearning
- Investigating Tool-Memory Conflicts in Tool-Augmented LLMs
- OneRec: Unifying Retrieve and Rank with Generative Recommender and Iterative Preference Alignment
- Who's Harry Potter? Approximate Unlearning in LLMs
- Towards LLM Unlearning Resilient to Relearning Attacks: A Sharpness-Aware Minimization Perspective and Beyond
- Erasing Conceptual Knowledge from Language Models
- Certified Data Removal from Machine Learning Models
- SOUL: Unlocking the Power of Second-Order Optimization for LLM Unlearning
- MiniOneRec: An Open-Source Framework for Scaling Generative Recommendation
- The WMDP Benchmark: Measuring and Reducing Malicious Use With Unlearning
- A Volume-Price-Adjusted MACD Trading Strategy with Sensitivity Calibration for U.S. Equity Indices
- Reflect-Guard: Enhancing LLM Safeguards against Adversarial Prompts via Logical Self-Reflection
- Rethinking Machine Unlearning for Large Language Models
- Improving the Completeness and Comparability of Segment Disclosures: A Large Language Model Approach
- OneRec-Think: In-Text Reasoning for Generative Recommendation
- Knowledge Conflicts for LLMs: A Survey
- Qwen3 Technical Report
- Carbon-Aware Compute--Power Scheduling for AI Data Centers with Microgrid Prosumer Operations
Related papers
- UniRank: Benchmarking Ranking Models for Unified Sequential Modeling and Feature Interaction
- Revisiting Text Ranking in Deep Research
- GateSID: One Gate to Balance Semantic and Collaborative Signals across the Item Lifecycle
- SAM-D2Q: Aligning Multimodal Doc2Query with Search Demand and Conversion for E-commerce
- GINGER: Spicing Up Graph-Based Vector Search
- Subtraction Gets You More: Gap-Aware Retrieval for Multimodal Multi-Hop QA