RGSD distills rubric-conditioned teacher distributions into base policies token-by-token, matching GRPO rubric satisfaction on Qwen models with one rollout and zero verifier calls.
hub
Writingbench: A comprehensive benchmark for generative writing
17 Pith papers cite this work. Polarity classification is still indexing.
17
Pith papers citing it
hub tools
citation-role summary
dataset 2
baseline 1
citation-polarity summary
representative citing papers
RDPO applies magnitude-aware quantile normalization and Mahalanobis whitening to decorrelate heterogeneous rewards in multi-objective RL, improving instruction following and writing quality on LongCat-Flash post-training while staying competitive on reasoning and coding.
LitVISTA benchmark shows frontier LLMs fail to jointly capture narrative function and structure in literary texts, with errors dominated by anchor identification.
A mutual evaluation system for LLMs that uses game-theoretic aggregation of peer reviews and validates alignment with human voting on subjective outputs.
MetaPS trains models via simulation rollouts to select from programmatic strategy libraries for market agents, yielding better performance than fixed or direct LLM baselines across model sizes.
NarrativeWorldBench evaluates 21 LLMs on nine narrative metrics across horizons to 200 episodes and introduces N-VSSM, a 256-dimensional variational state-space model that achieves plot-beat F1 >=0.84 with 4x lower compute and wins writer preference on consistency.
Later-domain RL training harms earlier domains via second-order damage concentrated in a low-dimensional shared conflict subspace; brief domain refresh contracts this component to enable selective recovery.
Rubric-based RL verifiers can be gamed via partial criterion satisfaction and implicit-to-explicit tricks, yielding proxy gains that do not improve quality under rubric-free judges; stronger verifiers reduce but do not eliminate the mismatch.
SEIF creates a self-reinforcing loop in which an LLM alternately generates increasingly difficult instructions and learns to follow them better using reinforcement learning signals from its own judgments.
LongWriter-Zero applies RL from a base model with specialized rewards for length, quality, and structure to outperform SFT baselines and larger models on long-writing benchmarks.
DeepResearch Bench supplies 100 expert-crafted PhD-level tasks and two human-aligned evaluation frameworks to measure deep research agents on report quality and citation accuracy.
MAGNET multi-agent generation with persona grounding and ATLAS graph verification yields 34-50% fewer hallucinations and annotations than single-model or IBSEN baselines at 100-page scale.
EngGPT2MoE-16B-A3B matches or exceeds other Italian open-source LLMs on most international benchmarks while remaining competitive on ITALIC, though it trails some top international models.
The paper introduces HarDBench, a benchmark demonstrating that LLMs can be jailbroken via draft-based co-authoring, and proposes a preference optimization method to mitigate this risk.
MindDR combines a Planning Agent, DeepSearch Agent, and Report Agent with SFT cold-start, Search-RL, Report-RL, and preference alignment to reach competitive scores on research benchmarks using 30B-scale models.
Pith review generated a malformed one-line summary.
citing papers explorer
-
Rubric-Guided Self-Distillation: Post-Training Without Rubric Verifiers
RGSD distills rubric-conditioned teacher distributions into base policies token-by-token, matching GRPO rubric satisfaction on Qwen models with one rollout and zero verifier calls.
-
Multi-Objective and Mixed-Reward Reinforcement Learning via Reward-Decorrelated Policy Optimization
RDPO applies magnitude-aware quantile normalization and Mahalanobis whitening to decorrelate heterogeneous rewards in multi-objective RL, improving instruction following and writing quality on LongCat-Flash post-training while staying competitive on reasoning and coding.
-
LitVISTA: A Benchmark for Narrative Orchestration in Literary Text
LitVISTA benchmark shows frontier LLMs fail to jointly capture narrative function and structure in literary texts, with errors dominated by anchor identification.
-
LLMs Judge Themselves: A Game-Theoretic Framework for Human-Aligned Evaluation
A mutual evaluation system for LLMs that uses game-theoretic aggregation of peer reviews and validates alignment with human voting on subjective outputs.
-
MetaPS: Adaptive Programmatic Strategy Selection for Market Agents
MetaPS trains models via simulation rollouts to select from programmatic strategy libraries for market agents, yielding better performance than fixed or direct LLM baselines across model sizes.
-
NarrativeWorldBench: A Frontier-Saturated Benchmark and a Latent World Model for Long-Horizon Co-Creative Audio Drama
NarrativeWorldBench evaluates 21 LLMs on nine narrative metrics across horizons to 200 episodes and introduces N-VSSM, a 256-dimensional variational state-space model that achieves plot-beat F1 >=0.84 with 4x lower compute and wins writer preference on consistency.
-
A Local Perturbation Theory for Cross-Domain Interference and Recovery in Multi-Domain RL
Later-domain RL training harms earlier domains via second-order damage concentrated in a low-dimensional shared conflict subspace; brief domain refresh contracts this component to enable selective recovery.
-
Reward Hacking in Rubric-Based Reinforcement Learning
Rubric-based RL verifiers can be gamed via partial criterion satisfaction and implicit-to-explicit tricks, yielding proxy gains that do not improve quality under rubric-free judges; stronger verifiers reduce but do not eliminate the mismatch.
-
LongWriter-Zero: Mastering Ultra-Long Text Generation via Reinforcement Learning
LongWriter-Zero applies RL from a base model with specialized rewards for length, quality, and structure to outperform SFT baselines and larger models on long-writing benchmarks.
-
From Personas to Plot: Character-Grounded Multi-Agent Story Generation for Long-Form Narratives
MAGNET multi-agent generation with persona grounding and ATLAS graph verification yields 34-50% fewer hallucinations and annotations than single-model or IBSEN baselines at 100-page scale.
-
Mind DeepResearch Technical Report
MindDR combines a Planning Agent, DeepSearch Agent, and Report Agent with SFT cold-start, Search-RL, Report-RL, and preference alignment to reach competitive scores on research benchmarks using 30B-scale models.
-
Qwen3 Technical Report
Pith review generated a malformed one-line summary.