# Enhancing Action and Ingredient Modeling for Semantically Grounded Recipe Generation

arXiv: https://arxiv.org/abs/2602.15862 · HTML: https://mlsanity.com/paper/2602.15862/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Placed in Image editing and quality assessment, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 20 of its 22 references that are in the map.
- First posted on arXiv 26 Jan 2026.

## Placed in

- Multimodal LLMs / vision-language models / [Image editing and quality assessment](https://mlsanity.com/mllm/image-editing-and-quality-assessment/)

## Builds on (most cited first, 20 of 20)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://mlsanity.com/paper/2005.11401/) (2020)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [GPT-4o System Card](https://mlsanity.com/paper/2410.21276/) (2024)
- [Improved Baselines with Visual Instruction Tuning](https://mlsanity.com/paper/2310.03744/) (2023)
- [Retrieval-Augmented Generation for Large Language Models: A Survey](https://mlsanity.com/paper/2312.10997/) (2023)
- [DeBERTa: Decoding-enhanced BERT with Disentangled Attention](https://mlsanity.com/paper/2006.03654/) (2020)
- [UniBind: LLM-Augmented Unified and Balanced Representation Space to Bind Them All](https://mlsanity.com/paper/2403.12532/) (2024)
- [Inverse Cooking: Recipe Generation from Food Images](https://mlsanity.com/paper/1812.06164/) (2018)
- [EVCap: Retrieval-Augmented Image Captioning with External Visual-Name Memory for Open-World Comprehension](https://mlsanity.com/paper/2311.15879/) (2023)
- [FoodLMM: A Versatile Food Assistant using Large Multi-modal Model](https://mlsanity.com/paper/2312.14991/) (2023)
- [Retrieval-augmented Image Captioning](https://mlsanity.com/paper/2302.08268/) (2023)
- [FIRE: Food Image to REcipe generation](https://mlsanity.com/paper/2308.14391/) (2023)
- [ChefFusion: Multimodal Foundation Model Integrating Recipe and Food Image Generation](https://mlsanity.com/paper/2409.12010/) (2024)
- [From Canteen Food to Daily Meals: Generalizing Food Recognition to More Practical Scenarios](https://mlsanity.com/paper/2403.07403/) (2024)
- [Navigating Weight Prediction with Diet Diary](https://mlsanity.com/paper/2408.05445/) (2024)
- [Retrieval Augmented Recipe Generation](https://mlsanity.com/paper/2411.08715/) (2024)

## Built on by (most cited first, 3 of 3)

- [SpatialImaginer: Towards Adaptive Visual Imagination for Spatial Reasoning](https://mlsanity.com/paper/2604.17385/) (2026)
- [GeoWeaver: Grounding Visual Tokens with Geometric Evidence before Scene Reasoning](https://mlsanity.com/paper/2605.22558/) (2026)
- [ReGraph: Learning to Generate Recipe Graphs from Food Images](https://mlsanity.com/paper/2608.06917/) (2026)

## Abstract (arXiv)

Recent advances in Multimodal Large Language Models (MLMMs) have enabled recipe generation from food images, yet outputs often contain semantically incorrect actions or ingredients despite high lexical scores (e.g., BLEU, ROUGE). To address this gap, we propose a semantically grounded framework that predicts and validates actions and ingredients as internal context for instruction generation. Our two-stage pipeline combines supervised fine-tuning (SFT) with reinforcement fine-tuning (RFT): SFT builds foundational accuracy using an Action-Reasoning dataset and ingredient corpus, while RFT employs frequency-aware rewards to improve long-tail action prediction and ingredient generalization. A Semantic Confidence Scoring and Rectification (SCSR) module further filters and corrects predictions. Experiments on Recipe1M show state-of-the-art performance and markedly improved semantic fidelity.