Persim Gemma 12b

--- license: apache-2.0 base_model: google/gemma-4-12B-it datasets: - xianyao/persim-sft language: - en library_name: transformers pipeline_tag: text-generation tags: - robotics - household - personality - layout-generation - gemma - lora --- # persim-gemma-12b An open-weight **Big-Five personality → household layout** generator for the [PerSim](https://github.com/XianyaoLi/PerSim) pipeline ([arXiv:2607.00022](https://arxiv.org/abs/2607.00022), IROS 2026). Given the pipeline's `generate_layouts.py` prompt (scene + persona + item vocabulary), it produces personality-conditioned placements as strict JSON: ```json { "room_movable_items": {"kitchen_0": ["mug_0", "plate_0", ...], ...}, "initial_object_positions": {"mug_0": {"relation": "on", "anchor_object": "countertop_tpuwys_0"}, ...} } ``` The pipeline works with any OpenAI-compatible endpoint; this model is a ready-made open option for the layout stage (the paper's fine-tuned Gemini 2.5 Flash cannot be redistributed). Outputs inherit only the pipeline's automatic checks — no plausibility or paper-equivalence claims. ## Training QLoRA (r=16) on `google/gemma-4-12B-it`, 2 epochs on one RTX 4090. Data: [persim-sft](https://huggingface.co/datasets/xianyao/persim-sft) human placement anchors (rigidity dropped) + layout demonstrations self-generated by the base model on the real pipeline prompts and filtered through `check_layouts.py`. No proprietary-model outputs in the supervision. ## Usage ```bash vllm serve xianyao/persim-gemma-12b \ --max-model-len 16384 \ --override-generation-config '{"repetition_penalty": 1.05}' ``` ```bash python generate_persona.py --model google/gemma-4-12B-it # base model python generate_layouts.py --model xianyao/persim-gemma-12b # this model python generate_trajectories.py --model google/gemma-4-12B-it # base model ``` Also answers the standalone anchor query (`Personality: O=..., C=...\nObject: mug` → `{"rooms": [...], "cooccurrence": [...]}`). ## Limitations - The largest shipped scene (`house_single_floor`, 120–240 items) is unreliable: generation can run away into numbered duplicates until the token limit. `repetition_penalty=1.05` plus the pipeline's retries partially compensate. Smaller scenes are stable. - Trained on the pipeline's five scenes and item vocabulary; new scenes are untested. - Layout plausibility is not validated — format and containment checks...

Source context: 925 downloads · 0 likes · Pipeline text-generation · Library transformers · Repo xianyao/persim-gemma-12b

Source context: 1287 downloads · 0 likes · Pipeline text-generation · Library transformers · Repo xianyao/persim-gemma-12b