A supervised fine-tuning approach using inverted multi-resolution planning scaffolds from public-domain novels trains models to generate book-length stories with more human-like literary qualities than standard instruction-tuned LLMs.
Parameterized Synthetic Text Generation with SimpleStories.arXiv preprint arXiv:2504.09184, 2025.https://arxiv.org/abs/2504.09184
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
citation-role summary
dataset 1
method 1
citation-polarity summary
years
2026 5verdicts
UNVERDICTED 5representative citing papers
LLMs perform in-context learning as trajectories through a structured low-dimensional conceptual belief space, with the structure visible in both behavior and internal representations and causally manipulable via interventions.
Annotation-anchored training reduces semantic diversity collapse in post-trained language models by a factor of six compared to standard supervised fine-tuning while preserving instruction-following and improving with scale.
Mixture-of-experts flow matching enables non-autoregressive language models to achieve autoregressive-level quality in three sampling steps, delivering up to 1000x faster inference than diffusion models.
Sessa integrates attention within recurrent paths to achieve power-law memory tails and flexible non-decaying selective retrieval, outperforming baselines on long-context tasks.
citing papers explorer
-
Towards Human-Level Book-Writing Capability
A supervised fine-tuning approach using inverted multi-resolution planning scaffolds from public-domain novels trains models to generate book-length stories with more human-like literary qualities than standard instruction-tuned LLMs.
-
Stories in Space: In-Context Learning Trajectories in Conceptual Belief Space
LLMs perform in-context learning as trajectories through a structured low-dimensional conceptual belief space, with the structure visible in both behavior and internal representations and causally manipulable via interventions.
-
Annotations Mitigate Post-Training Mode Collapse
Annotation-anchored training reduces semantic diversity collapse in post-trained language models by a factor of six compared to standard supervised fine-tuning while preserving instruction-following and improving with scale.
-
Towards Faster Language Model Inference Using Mixture-of-Experts Flow Matching
Mixture-of-experts flow matching enables non-autoregressive language models to achieve autoregressive-level quality in three sampling steps, delivering up to 1000x faster inference than diffusion models.
-
Sessa: Selective State Space Attention
Sessa integrates attention within recurrent paths to achieve power-law memory tails and flexible non-decaying selective retrieval, outperforming baselines on long-context tasks.