SatDiff-LFM2.5-VL-450M-stage1
A LoRA fine-tune of LiquidAI/LFM2.5-VL-450M trained to emit per-claim, contract-framed evidence on Sentinel-2 imagery of regulated dams and tailings storage facilities. Built for the SatDiff submission to the Liquid AI "AI in Space" hackathon (Liquid Track, 2026).
This is the evidence-text writer component of the SatDiff pipeline. It does not perform threshold-based severity classification on its own — that work is done deterministically by a Python rules engine downstream of the model. See Stage 2 (negative result) below for the methodological reasoning.
What it does
Given (a) Sentinel-2 imagery (RGB + NIR composites for baseline + current pass) and (b) a contract memo prompt that lists per-claim evidence-sourcing rules, this model produces a structured per-claim evidence string for a 5-claim audit schema covering:
- Impoundment morphology (deposition asymmetry, footprint change)
- Pond management (pond-to-wall distance, area change, turbidity, NDWI)
- Retaining-wall integrity (gully count + width, NDMI on the wall face, SWIR anomaly)
- Deformation (declares "no SAR data available" when SAR is absent)
- Protected-zone encroachment (towns, residential extensions)
Headline result (held-out evaluation)
The base LFM2.5-VL-450M parrots the same three indices into every claim's evidence regardless of which physical signal the claim is about. Stage 1 fine-tuning — without any SatDiff-specific examples — teaches the model to read the per-claim sourcing rules from the prompt and cite the right diff fields.
| metric (held-out backtest passes, 6 dates × 5 claims = 30 claims) | base | Stage 1 |
|---|
| schema-valid passes | 6/6 | 6/6 |
| evidence-correct claims | 0/30 (0 %) | 30/30 (100 %) |
The lift comes from generic VRSBench grounding, not from domain-specific examples.
Training
- Framework:
Liquid4All/leap-finetune (Ray Train + Accelerate, managed via uv). Not raw transformers + peft.
- Base model:
LiquidAI/LFM2.5-VL-450M.
- Dataset: VRSBench (NeurIPS 2024) — 5 000 captioning + VQA samples (no
[refer] grounding tasks).
- Method: LoRA SFT, rank as configured in the recipe, 2 epochs.
- Hardware: single RTX 4080 Laptop, 12 GB VRAM, WSL2 + CUDA 12.6.
- Wall-clock: 38 m 15 s.
- Eval loss: base ~3.21 → Stage 1 1.41 (−56 %).
Stage 2 (negative result, not shipped)
A second-stage fine-tune was attempted using 29 hand-authored examples (boundary, escalation, routine, catastrophic regimes) plus 17 auto-generated examples from the SatDiff Phase 2 backtest, split 35 train / 11 held-out before training. Stage 2 preserved Stage 1's 100 % evidence-correctness but worsened severity adherence on real held-out data (5 → 9 corrections by the downstream rules engine).
Diagnosis: the hand-authored cases used contrived metric values around the threshold cliffs (e.g. pond-to-wall = 24 m vs 26 m); the held-out real-data passes lived in a different distribution (pond-to-wall ≈ 9.9 m throughout the failure window). LoRA at this scale (35 examples × 3 epochs ≈ 31 effective steps) cannot reshape multi-tier threshold reasoning.
We ship Stage 1, not Stage 2. The Stage 2 checkpoint is intentionally not uploaded — it would confuse the model-card story.
This is the strongest possible validation of the SatDiff rules-engine architecture: severity, action, escalation, and overall-status are computed deterministically by phase2.aggregate.compute_severity from physical-diff numbers, regardless of what the model emits. Stage 2 attempting and failing to lift the model's threshold reasoning confirms that this work belongs in deterministic Python at our scale.
Files in this repo
- fp16 transformers checkpoint (
model.safetensors + config.json + tokenizer.json + chat_template.jinja + …) — load via transformers.AutoModelForImageTextToText.
- GGUF pair (
gguf/LFM2.5-VL-450M-stage1-Q8_0.gguf + gguf/mmproj-LFM2.5-VL-450M-stage1-Q8_0.gguf) — load via llama.cpp / llama-server. 362 MB Q8_0 backbone + 182 MB F16 mmproj = 544 MB total.
Reproducing the headline result
1# 1. Pull the SatDiff repo + Stage 1 GGUF, bring up the stack:
2git clone <satdiff-repo>
3cd satdiff
4docker compose up -d
5
6# 2. Run the Phase 2 backtest against the Stage 1 model:
7python -m phase2.cli --asset jagersfontein \
8 --date-range 2021-06-15,2022-10-15 \
9 --inference llama_server \
10 --model WobblyDopamine/SatDiff-LFM2.5-VL-450M-stage1-Q8_0
11
12# 3. Render the per-pass PDF audit reports:
13python -m phase3 --asset jagersfontein --date-range 2021-06-15,2022-10-15
End-to-end latency on RTX 4080 Laptop (sm_89, CUDA 12.6) using the GGUF + llama-server path: 2.38 s/pass — 4.7× over the bf16 transformers path with no schema or evidence-quality regression.
Intended use
This model is the evidence-text writer for the SatDiff TSF/dam compliance pipeline. It is not a general-purpose VLM and is not a severity classifier on its own. Use it as part of the rules-engine pipeline described above.
Limitations
- Trained on a small VRSBench slice; performance outside the SatDiff contract prompt's sourcing rules is unknown.
- Held-out evaluation was on 6 Jagersfontein passes; transfer to other TSF/dam assets is plausible (the lift comes from generic EO grounding) but unmeasured.
- Severity grading is performed by a downstream Python rules engine, not by this model. Do not delegate threshold reasoning to the fine-tuned weights at this scale.
- Cloud-cover gating is upstream; the model is not robust to severe cloud occlusion.
License
This work is released under the
LFM Open License v1.0, inherited from the base model
LiquidAI/LFM2.5-VL-450M. See the upstream license at
https://huggingface.co/LiquidAI/LFM2.5-VL-450M/blob/main/LICENSE.
Citation
1@misc{satdiff2026,
2 title={SatDiff: A Satellite-Readable Compliance Contract for Tailings and Dam Monitoring},
3 author={Scholz, Peter},
4 year={2026},
5 howpublished={Liquid AI "AI in Space" Hackathon submission, Liquid Track},
6 note={Fine-tune of LiquidAI/LFM2.5-VL-450M on VRSBench. Stage 1 evidence-correctness 0/30 → 30/30 on held-out tailings-dam passes. See https://huggingface.co/WobblyDopamine/SatDiff-LFM2.5-VL-450M-stage1.}
7}
Acknowledgements
- Liquid AI for the LFM2.5-VL-450M base model and the
leap-finetune framework.
- DPhi for the SimSat API and the hackathon platform.
- Torres-Cruz & O'Donovan (2023) for the Scientific Reports reconstruction of the Jagersfontein failure that anchored this submission's validation.
