Liquid AI Open-Sources Antidoom: A Final Token Preference Optimization (FTPO) Method that Reduces Doom Loops in Reasoning Models
Liquid AI has released Antidoom, an open-source method that targets a common failure mode in reasoning models. That failure mode is the doom loop. In a doom loop, a model emits a span. It then repeats that span again and again. The output continues until the context window is exhausted. Small reasoning models are more prone to this, especially on long thinking traces and hard problems. On an early checkpoint of LFM2.5-2.6B, 10.2% of completions on hard math and coding prompts produced repeti
Get smart on it
Reasoning models sometimes get stuck in "doom loops," where they repeat the same span of text over and over until running out of space. A method called Antidoom addresses this by identifying the specific token that starts each loop and retraining the model to prefer alternative tokens at that position instead. The approach, which uses Final Token Preference Optimization (FTPO), does not teach the model new information but rather clears repetitive outputs that were blocking answers the model could already produce. Testing showed this reduced looping from 10.2% to 1.4% on one model, with improvements in evaluation scores across the board.
Related news
Ant Group’s Robbyant Open-Sources LingBot-Vision: A 1B Boundary-Centric Vision Foundation Model for Dense Spatial Perception
Robbyant, the embodied-AI company within Ant Group, has open-sourced LingBot-Vision, a family of self-supervised Vision Transformers built for dense spatial perception. The weights ship under Apache-2.0 on Hugging Face in four sizes — ViT-giant, ViT-large, ViT-base, and ViT-small — together with a technical report and inference code. Most vision foundation models are trained for semantic invariance: they learn to answer what is in an image while discarding exactly the fine-grained spatial st
NVIDIA Releases Audex (Nemotron-Labs-Audex-30B-A3B): A Unified Audio-Text LLM That Preserves the Text Intelligence of Its Backbone
NVIDIA has released Audex (Nemotron-Labs-Audex-30B-A3B), a unified audio-text large language model. It understands and generates both audio and speech. It also keeps the text intelligence of its backbone. The checkpoints, along with a smaller Audex-2B, are released under a noncommercial license. Most multimodal models pay a text tax. When labs add audio or vision output, text benchmarks often drop. NVIDIA research team reports this even for speech-only output models. Audex is designed to avo
