javadtaghia/unlearning-via-weight-code
Updated
Javad Taghia | PhD PRO
javadtaghia
AI & ML interests
Work with agentic systems, agent memory, model fine tuning and unlearning, generative AI, and the development and deployment of open source AI
Recent Activity
commentedon their article about 3 hours ago
An LLM erases a concept by editing diffusion weights directly, without gradients updated a Space about 6 hours ago
javadtaghia/ai-written-rank posted an update about 7 hours ago
Can a general-purpose LLM erase a concept from a diffusion model by editing its weights directly, with no gradients?
We tested it under a preregistered protocol. GPT-6 Astra got tool access to the weights of segmind/tiny-sd and had to remove Van Gogh's painting style. Every edit was a closed-form tensor operation in a sandbox that blocks all autograd paths.
Held-out test, confirmatory runs (paired, template-clustered 95% CIs):
• vs random search with the same edit space and the same feedback: −52.1 pts Van Gogh rate [−61.7, −42.3]
• vs UCE (closed-form erasure): −11.5 pts [−22.9, −0.9], with identical damage to neighbouring painters
• vs ESD (gradient fine-tuning): equal removal, +11.8 pts better neighbour preservation
Limits, stated up front: one model × one concept so far; no method (the agent included) removes the style when it is described without the artist's name; no recovery attacks yet.
Every run is public: hash-chained logs, budgets, the agent's full transcripts, per-prompt scores, checkpoints, the frozen protocol and costs.
Article: https://huggingface.co/blog/javadtaghia/an-llm-erases-a-concept-by-editing-diffusion-weigh
Data: https://huggingface.co/datasets/javadtaghia/unlearning-via-weight-public
Code: https://huggingface.co/javadtaghia/unlearning-via-weight-code
Organizations
javadtaghia/esd_NSFW
Updated
javadtaghia/uce_NSFW
Updated
javadtaghia/dee-z-image
Text-to-Image • 6B • Updated • 16 • 1
javadtaghia/dee-unlearning-tiny-sd
Text-to-Image • Updated • 19 • 1
javadtaghia/deewaiREALCN
Text-to-Image • 3B • Updated • 7 • 2
javadtaghia/dee-tulu-train
Text Generation • Updated