Learning the Neighborhood: Contrast-Free Multimodal Self-Supervised Molecular Graph Pretraining

Kari Jaaskelainen

29 Sep 2025 — 1 min read

C-FREE fuses 2D graphs with 3D conformers using ego-nets—no negatives required

High-quality molecular representations often require large labeled datasets or fragile contrastive schemes. C-FREE offers a simpler, more powerful path: contrast-free, multimodal pretraining on both 2D topology and ensembles of 3D conformers. The core idea is to predict subgraph (ego-net) embeddings from their complementary neighborhoods in latent space, encouraging models to capture the mutual information between local structure and its broader molecular context—without negatives, positional encodings, or heavy preprocessing.

Technically, C-FREE uses fixed-radius ego-nets as consistent modeling units across conformers and integrates geometric and topological cues with a hybrid GNN–Transformer backbone. Pretrained on GEOM, a dataset rich in conformational diversity, it achieves state-of-the-art results on MoleculeNet, outperforming contrastive, generative, and other multimodal self-supervised methods.

Contrast-free objective: avoids negative sampling pitfalls.
Multimodal fusion: unifies 2D graphs and 3D conformers seamlessly.
Ego-net prediction: learns from complementary neighborhoods to encode context.
Simple and efficient: no positional encodings or expensive preprocessing.

Beyond benchmarks, fine-tuning on diverse datasets shows strong transfer to new chemical domains and molecule sizes. The message is clear: when 3D information is plentiful, pretraining strategies that learn the neighborhood—rather than contrast it—can unlock robust, generalizable molecular representations for property prediction, design, and discovery.

Paper: http://arxiv.org/abs/2509.22468v1
Register: https://www.AiFeta.com

#GraphML #GNN #ChemInformatics #SelfSupervised #MoleculeNet #3DConformers #RepresentationLearning

From f(x) and g(x) to f(g(x)): LLMs Learn New Skills in RL by Composing Old Ones

Evidence that RL teaches genuinely new abilities: compositional skills emerge and transfer across tasks Does RL merely reweight what an LLM already knows—or can it teach genuinely new skills? This paper offers concrete evidence for the latter. Using a controlled, synthetic framework, the authors define “skills” as string transformation

OpenGPT-4o-Image: A Comprehensive Dataset for Advanced Image Generation and Editing

A structured 80k instruction–image corpus spanning 11 domains and 51 subtasks to train unified visual editors Unified models for image generation and editing hit a data ceiling: existing corpora emphasize basic manipulations but miss real‑world complexity. OpenGPT‑4o‑Image tackles this with a hierarchical task taxonomy and automated

Random Policy Valuation is Enough for LLM Reasoning with Verifiable Rewards

ROVER replaces PPO loops with uniform‑policy Q‑values—boosting quality and diversity in math reasoning Popular RLVR methods for LLM reasoning lean on generalized policy iteration (e.g., PPO/GRPO), but suffer instability and diversity collapse. This paper reframes math RLVR as a specialized finite‑horizon MDP with deterministic

CLPO: Curriculum Learning meets Policy Optimization for LLM Reasoning

A dynamic, self‑paced curriculum that restructures problems to match model ability in RLVR Online RL with Verifiable Rewards (RLVR) has boosted LLM reasoning—but most methods treat all problems equally, wasting effort on solved items and flailing on those beyond current capability. CLPO fixes that with a dynamic pedagogy:

Read more

From f(x) and g(x) to f(g(x)): LLMs Learn New Skills in RL by Composing Old Ones

OpenGPT-4o-Image: A Comprehensive Dataset for Advanced Image Generation and Editing

Random Policy Valuation is Enough for LLM Reasoning with Verifiable Rewards

CLPO: Curriculum Learning meets Policy Optimization for LLM Reasoning