QVal is a new evaluation framework that directly measures dense supervision quality via Q-alignment to a reference policy, showing simple prompting baselines outperform 21 other methods across environments and models.
When to solve, when to verify: Compute-optimal problem solving and generative verification for llm reasoning
7 Pith papers cite this work. Polarity classification is still indexing.
7
Pith papers citing it
citation-role summary
background 2
citation-polarity summary
roles
background 2representative citing papers
Preregistered placebo-controlled decomposition shows external executable counterexamples drive self-repair gains in small code models more than re-exposure or self-critique.
MCMit proposes a constant-latency multi-control branch instruction, transformer and CNN discriminators, plus static MCM elimination and stochastic branching, evaluated on Qubic with QPU traces to cut latency by 70% and logical error rates by up to 9.4x.
SIEVES improves selective prediction coverage by up to 3x on OOD VQA benchmarks by training a selector to score the quality of visual evidence produced by reasoner models, generalizing across benchmarks and proprietary models without internal access or per-task retraining.
Fine-tuned LLM judges struggle with future-proofing to newer generators but maintain backward-compatibility more easily; DPO training and continual learning improve adaptation while all models degrade on unseen questions.
Iterative SFT-RL cycles enable a 7B LVLM to develop sophisticated visual chain-of-thought reasoning and improve performance on math and general reasoning benchmarks.
LLM reasoning is primarily mediated by latent-state trajectories rather than by explicit surface chain-of-thought outputs.
citing papers explorer
-
QVal: Cheaply Evaluating Dense Supervision Signals for Long-Horizon LLM Agents
QVal is a new evaluation framework that directly measures dense supervision quality via Q-alignment to a reference policy, showing simple prompting baselines outperform 21 other methods across environments and models.
-
Falsification, Not Exposure: An Internally Preregistered Placebo-Controlled Decomposition of Self-Repair Feedback in Frozen Small Code Models
Preregistered placebo-controlled decomposition shows external executable counterexamples drive self-repair gains in small code models more than re-exposure or self-critique.
-
MCMit: Mid-Circuit Measurement Error Mitigation
MCMit proposes a constant-latency multi-control branch instruction, transformer and CNN discriminators, plus static MCM elimination and stochastic branching, evaluated on Qubic with QPU traces to cut latency by 70% and logical error rates by up to 9.4x.
-
SIEVES: Selective Prediction Generalizes through Visual Evidence Scoring
SIEVES improves selective prediction coverage by up to 3x on OOD VQA benchmarks by training a selector to score the quality of visual evidence produced by reasoner models, generalizing across benchmarks and proprietary models without internal access or per-task retraining.
-
On the Shelf Life of Fine-Tuned LLM-Judges: Future-Proofing, Backward-Compatibility, and Question Generalization
Fine-tuned LLM judges struggle with future-proofing to newer generators but maintain backward-compatibility more easily; DPO training and continual learning improve adaptation while all models degrade on unseen questions.
-
OpenVLThinker: Complex Vision-Language Reasoning via Iterative SFT-RL Cycles
Iterative SFT-RL cycles enable a 7B LVLM to develop sophisticated visual chain-of-thought reasoning and improve performance on math and general reasoning benchmarks.
-
LLM Reasoning Is Latent, Not the Chain of Thought
LLM reasoning is primarily mediated by latent-state trajectories rather than by explicit surface chain-of-thought outputs.