TwIL-LM3

TwIL-LM3

A 3B reasoning model for formal logic tasks, built from
`HuggingFaceTB/SmolLM3-3B` through LoRA
supervised fine-tuning, checkpoint fusion, WiSE-FT weight interpolation, and entropy-weighted
GRPO reinforcement learning.

It improves in-domain formal-logic performance by +26% relative over its base model
(macro gate 0.336 → 0.422) and improves held-out benchmark performance at the same time
(+0.022 core average). It is the only arm in this project that gains on both tracks, which is
why it is the recommended release of the pair.

Try out the model on our TwIL-LM3.1 branch, a better version of TwIL-LM3.

Highlights

Do check out our new model TwIL-LM3-Pro, a much better version of the TwIL-LM3 at 3.66B parameters.

• Gains on both tracks at once — the only arm in this project that does. In-domain macro gate
• 0.336 → 0.422, and the held-out 10-dataset macro 0.7193 → 0.7339 rather than the usual collapse
• that follows task-specific fine-tuning.
• Beats every arm up to and including LFM2.5-8B-A1B — roughly three times its parameter count
• — on all six Track A objective lanes and all four summary rows, not on average alone.
• Competitive with 8B on strict scoring. On strict-7, which gives no loose-match credit
• anywhere, it sits 0.012 behind Qwen3-8B (0.1971 against 0.2093) at 2.6x fewer parameters, and
• ahead of it on Lean formalisation (token-F1 0.5869 against 0.4022) and semantic parsing (0.4416
• against 0.4257).
• Structured formal output. Tuned for the objects rather than the prose: FOL translation,
• entailment labels, semantic parses, Lean statements and Lean proof critique.
• The most efficient arm measured, at any scale. 482-token Track B generations and 32.9
• completed answers per second — about eight times gpt-oss-120b's rate — because it answe…

Source: https://huggingface.co/webAI-Official/TwIL-LM3