Weak-to-strong generalization is nearly inevitable in linear logistic regression for most student-teacher pairs without any model capacity mismatch.
hub Canonical reference
CoRR , volume =
Canonical reference. 88% of citing Pith papers cite this work as background.
25
Pith papers citing it
Background
88% of classified citations
hub tools
citation-role summary
background 7
method 1
citation-polarity summary
representative citing papers
KAT detects persistent low-KL agreement traps in on-policy distillation via a dynamic threshold to filter weak supervision, improving avg@k by 2.66% and pass@k by 3.43% on four math benchmarks while shortening rollouts by 59.73%.
S-GRPO unifies SFT and RL for LVLMs via conditional ground-truth injection that supplies a maximal-reward anchor when group exploration fails completely.
GPRL carries a k-dimensional skew-symmetric preference structure into policy updates with per-dimension advantages and a drift monitor, yielding 56.51% length-controlled win rate on AlpacaEval 2.0 from Llama-3-8B-Instruct while outperforming SimPO and SPPO on other benchmarks.
Forgetting in LLM continual post-training is a geometry conflict between task-induced covariance structures and the evolving model state, controlled by gating Wasserstein barycenter merging on measured conflict.
CLR-voyance reformulates inpatient reasoning as POMDP with clinician-validated outcome rubrics, yielding an 8B model that outperforms larger frontier models on the authors' new benchmark.
Perturbing prefixes to semantic neighbors during training creates a hierarchical noise model that improves language model predictions on token sequences outside the training corpus support.
Non-model gains via inference, systems, and assets can drive AI capabilities independently of base models, requiring governance beyond model-level evaluation and mitigation.
Recovering an orthogonal basis from model activations yields a model-native skill characterization that improves reasoning Pass@1 by up to 41% via targeted data selection and supports inference steering, outperforming human-characterized alternatives.
CoME-VL fuses contrastive and self-supervised vision encoders via entropy-guided multi-layer aggregation and RoPE cross-attention to improve vision-language model performance on benchmarks.
ETS performs training-free RL alignment for language models by energy-guided test-time scaling with Monte Carlo energy estimation and importance sampling acceleration.
SPaCe uses semantic clustering to shrink training sets and a multi-armed bandit to adaptively select samples, matching or beating baselines on reasoning benchmarks with up to 100x fewer examples.
A principled reward design for tool selection and application in RL-trained LLMs delivers 17% gains over base models and 15% over SFT across benchmarks.
PriFT uses token reweighting signals from a frozen pretrained model to stabilize SFT and achieve better results than standard SFT baselines on reasoning tasks.
Thinking-RFT improves Theory of Mind accuracy by 6% over SFT on shortcut-free datasets, with 10% gains on higher-order reasoning and better generalization to new domains.
IMAX trains soft prefixes with an InfoMax reward to drive diverse exploration in RLVR, yielding up to 11.60% gains in Pass@4 over standard RLVR across model scales.
AI data firms view human expertise as an extractable, low-cost resource to feed AI systems while treating institutional expertise as something needing liberation or reform to fit this model.
CPO++ adapts reinforcement fine-tuning of MLLMs to endogenous multi-modal concept drift through counterfactual reasoning and preference optimization, yielding better coherence and cross-domain robustness in safety-critical settings.
SignReasoner decomposes traffic signs into functional structure units and uses a two-stage VLM post-training pipeline to achieve state-of-the-art compositional reasoning on a new benchmark.
LLM post-training is unified as off-policy or on-policy interventions that expand support for useful behaviors, reshape policies within reachable states, or consolidate behavior across training stages.
ProxyCoT transfers CoT reasoning from proxy short contexts to full long contexts through RL/distillation followed by SFT, outperforming baselines with lower overhead and generalizing out-of-domain.
LoRR augments preference optimization methods like DPO with high-replay training, periodic resets to initial data/policy, and a hybrid objective to improve sample efficiency and reduce primacy bias on math and reasoning tasks.
Libra optimizes GPU allocation across rollout and training in agentic RL via an elastic hybrid pool and C-MLFQ scheduler based on tool-return causal signals, claiming up to 3.0x throughput and 2.5x faster reward convergence on 48 A800 GPUs.
Reinforcement learning is advanced for communication-efficient federated optimization and for preference-aligned, contextually safe policies in large language models.
citing papers explorer
No citing papers match the current filters.