Crimson AI NewsA CrimsonLingua Network service
EN ع
← Back to news
Research paper Hugging Face

Small Cognitive Models Match Giants In-Distribution but Scale Better Out-of-Distribution

AI By Crimson AI Hugging Face Papers 10 August 2026 · 00:00 8 views
Share: X Telegram

New research shows that small language models fine-tuned on human behavioral data can match a 70B baseline in-distribution, but larger models generalize better to novel tasks. The models rely on task stimuli and feedback, not just choice history.

Small Cognitive Models Match Giants In-Distribution but Scale Better Out-of-Distribution

Key points

A new paper from Hugging Face researchers investigates whether small language models can serve as cognitive proxies for human behavior. The team trained 14 models ranging from 135M to 14B parameters across four architecture families on Psych-101, a dataset of 10.7 million trial-level choices from 160 experiments.

Surprisingly, for in-distribution simulations, scale barely matters. Models from 0.6B to 1B parameters match a 70B baseline on held-out participants, falling within a narrow performance band. However, out-of-distribution, the band widens into a steeper scaling gradient, with larger models clearly advantaged in generalizing to novel task structures.

To understand what information the models use, the researchers ran diagnostics by progressively stripping four prompt channels—task instructions, experimental stimuli, outcome feedback, and choice history—across 27 experiments, and permuting trial order. Masking stimuli and feedback destroyed 75.7% of learned information and pushed models below chance, demonstrating that choice history alone does not account for performance.

Permutation revealed invariance on tasks with independent trials but sensitivity where trial order is determined by prior responses. The authors conclude that small cognitively fine-tuned models show promise as noise ceiling estimators for psychological experiments, though their scope remains bounded by training paradigms.

Model FamilySizes (Parameters)LoRA Ranks Swept
Qwen3-Base0.6B, 1.7B, 4B, 8B, 14Br=4,8,16,32,64 (except Qwen3-1.7B: r=8,16; Qwen3-14B: r=4,16,64)
Llama-31B, 3B, 8Br=4,8,16,32,64
SmolLM135M, 360M, 1.7B (SmolLM2), 3B (SmolLM3)r=4,8,16,32,64
OLMo1B (OLMo-2), 7B (OLMo-3)r=4,8,16,32,64
Source
Hugging Face · Hugging Face Papers
Related news
Research paper
Hugging Face 31 Aug 2026

Hugging Face Unveils StepGuard: Step-Level Guardrails for Safer AI Agents

StepGuard, a new step-level guard model from Hugging Face, audits agent actions before execution, reducing attack success rates by...

1
Research paper
Hugging Face 31 Aug 2026

Hugging Face Researchers Unveil ABot-Recon for Stable Long-Horizon 3D Reconstruction

ABot-Recon, a new streaming 3D reconstruction model from Hugging Face, achieves stable long-horizon performance using only local t...

1
Research paper
Hugging Face 31 Aug 2026

ContextPilot: Teaching Agents Proactive Context Management via Fine-Grained RL

Hugging Face researchers introduce ContextPilot, a framework that enhances long-horizon agent reasoning by expanding context-editi...

1