FlexiSLM is the first spoken language model supporting dynamic and controllable frame rates on speech input and output, outperforming fixed-rate 7B models at high quality and enabling faster inference at lower rates like 6.25 Hz.
Stay on topic with classifier-free guidance
8 Pith papers cite this work, alongside 5 external citations. Polarity classification is still indexing.
8
Pith papers citing it
5
external citations · external index
years
2026 8verdicts
UNVERDICTED 8representative citing papers
EGPS localizes MCMC moves to high-entropy decision points using forward-pass entropy, yielding up to 12.6× wall-clock speedup and best-or-tied accuracy on MATH500, HumanEval, and GPQA for Qwen2.5-Math-7B.
Inducing artificial uncertainty on trivial tasks allows training probes that achieve higher calibration on hard data than standard approaches while retaining performance on easy data.
GCPO performs per-token credit assignment in discrete policy optimization by setting token advantages proportional to the difference in model predictions under positive versus negative prompts, outperforming GRPO and DAPO on text-to-image and chain-of-thought tasks.
EmoMind is the first end-to-end pipeline that decodes continuous affective captions from fMRI by combining brain-decoded visual features with a 34D emotion vector and classifier-free guidance to balance semantic fidelity and affective expressivity.
A latent mediation framework with sparse autoencoders enables non-additive token-level influence attribution in LLMs by learning orthogonal features and back-propagating attributions.
A GPT2 model with discrete property tokens is fine-tuned via multi-task learning to generate molecules whose TDDFT-computed optical properties match conditioning targets, with controllability varying by chemical motif.
SPREG detects logical failures in LLM long-chain reasoning through real-time entropy spikes and performs structured plan repairs using historical distributions, reporting a 20% absolute accuracy gain on AIME25.
citing papers explorer
-
FlexiSLM: A Dynamic and Controllable Frame Rate Spoken Language Model
FlexiSLM is the first spoken language model supporting dynamic and controllable frame rates on speech input and output, outperforming fixed-rate 7B models at high quality and enabling faster inference at lower rates like 6.25 Hz.
-
Sample Where You Struggle: Sharpening Base Model Reasoning via Entropy-Guided Power Sampling
EGPS localizes MCMC moves to high-entropy decision points using forward-pass entropy, yielding up to 12.6× wall-clock speedup and best-or-tied accuracy on MATH500, HumanEval, and GPQA for Qwen2.5-Math-7B.
-
Inducing Artificial Uncertainty in Language Models
Inducing artificial uncertainty on trivial tasks allows training probes that achieve higher calibration on hard data than standard approaches while retaining performance on easy data.
-
Guidance Contrastive Token Credit Assignment for Discrete Policy Optimization
GCPO performs per-token credit assignment in discrete policy optimization by setting token advantages proportional to the difference in model predictions under positive versus negative prompts, outperforming GRPO and DAPO on text-to-image and chain-of-thought tasks.
-
EmoMind: Decoding Affective Captions from Human Brain fMRI
EmoMind is the first end-to-end pipeline that decodes continuous affective captions from fMRI by combining brain-decoded visual features with a 34D emotion vector and classifier-free guidance to balance semantic fidelity and affective expressivity.
-
Correcting Influence: Unboxing LLM Outputs with Orthogonal Latent Spaces
A latent mediation framework with sparse autoencoders enables non-additive token-level influence attribution in LLMs by learning orthogonal features and back-propagating attributions.
-
De novo molecular generation with optical property preconditioning at the token level
A GPT2 model with discrete property tokens is fine-tuned via multi-task learning to generate molecules whose TDDFT-computed optical properties match conditioning targets, with controllability varying by chemical motif.
-
SPREG: Structured Plan Repair with Entropy-Guided Test-Time Intervention for Large Language Model Reasoning
SPREG detects logical failures in LLM long-chain reasoning through real-time entropy spikes and performs structured plan repairs using historical distributions, reporting a 20% absolute accuracy gain on AIME25.