Pith. sign in

REVIEW 16 cited by

NoveltyBench: Evaluating Language Models for Humanlike Diversity

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2504.05228 v4 pith:KR52H6MG submitted 2025-04-07 cs.CL

NoveltyBench: Evaluating Language Models for Humanlike Diversity

classification cs.CL
keywords diversitymodelslanguagenoveltybenchbenchmarkscurrentdiverseelicit
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

Language models have demonstrated remarkable capabilities on standard benchmarks, yet they struggle increasingly from mode collapse, the inability to generate diverse and novel outputs. Our work introduces NoveltyBench, a benchmark specifically designed to evaluate the ability of language models to produce multiple distinct and high-quality outputs. NoveltyBench utilizes prompts curated to elicit diverse answers and filtered real-world user queries. Evaluating 20 leading language models, we find that current state-of-the-art systems generate significantly less diversity than human writers. Notably, larger models within a family often exhibit less diversity than their smaller counterparts, challenging the notion that capability on standard benchmarks translates directly to generative utility. While prompting strategies like in-context regeneration can elicit diversity, our findings highlight a fundamental lack of distributional diversity in current models, reducing their utility for users seeking varied responses and suggesting the need for new training and evaluation paradigms that prioritize diversity alongside quality.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 16 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Recalling Too Well: Sycophancy Evaluation and Mitigation in Memory-Augmented Models

    cs.AI 2026-06 conditional novelty 8.0

    Memory augmentation in LLMs amplifies sycophancy up to 25x compared to in-context baselines due to lossy memory extraction, with two lightweight mitigations that reduce the effect while preserving recall.

  2. Improving LLMs via Validator-to-Generator Alignment

    cs.CL 2026-07 conditional novelty 7.0

    Frequency-corrected rank alignment of an LLM generator to its own validator improves generator AUROC and G-V Pearson correlation by up to 27 points while preserving validator quality.

  3. AI Coding Agents in Social Science: Methodologically Diverse, Empirically Consistent, Interpretively Vulnerable

    cs.CL 2026-06 unverdicted novelty 7.0

    LLM agents match or exceed human methodological diversity and produce aligned effect estimates, yet flip final verdicts from 10% to 90% support under a confirmatory prompt while leaving coefficients unchanged.

  4. Assessing the Creativity of Large Language Models: Testing, Limits, and New Frontiers

    cs.AI 2026-05 conditional novelty 7.0

    The Divergent Remote Association Test (DRAT) is the first creativity test that significantly predicts LLMs' scientific ideation ability, unlike prior tests such as DAT or RAT.

  5. Ex Ante Evaluation of AI-Induced Idea Diversity Collapse

    cs.AI 2026-05 unverdicted novelty 7.0

    Frontier LLMs generate creative ideas with excess population-level crowding below human-relative parity across tasks, but targeted generation protocols can reduce it.

  6. Beyond One Output: Visualizing and Comparing Distributions of Language Model Generations

    cs.AI 2026-04 conditional novelty 7.0

    GROVE visualizes distributions of language model generations as overlapping paths through a text graph, with user studies showing that graph summaries aid structural judgments like diversity assessment while raw outpu...

  7. Beyond Semantic Manipulation: Token-Space Attacks on Reward Models

    cs.LG 2026-04 unverdicted novelty 7.0

    TOMPA performs black-box adversarial optimization in token space to discover non-linguistic patterns that nearly double the reward scores of GPT-5 answers on Skywork-Reward-V2 while producing gibberish text.

  8. On-Policy Self-Distillation with Sampled Demonstrations Reduces Output Diversity

    cs.LG 2026-06 unverdicted novelty 6.0

    On-policy self-distillation with sampled demonstrations reduces rollout diversity by amplifying existing probability gaps in the base model, unlike ideal RL which preserves ratios among correct outputs.

  9. Beyond Alignment: Value Diversity as a Collective Property in Multicultural Agent Systems

    cs.CL 2026-06 unverdicted novelty 6.0

    Multicultural multi-agent LLM systems exhibit substantially lower value diversity than human societies on the World Values Survey, with diversity uncorrelated to per-agent alignment and further reduced by agent interactions.

  10. "I've Seen How This Goes": Characterizing Diversity via Progressive Conditional Surprise

    cs.CL 2026-06 unverdicted novelty 6.0

    Decan (D_Ca_n = C × a_n) measures text diversity as progressive conditional surprise from base LM log-probabilities, scoring 0.846 OCA on McDiv benchmark and detecting monotonic diversity drop across base→SFT→DPO→RLVR stages.

  11. Annotations Mitigate Post-Training Mode Collapse

    cs.CL 2026-05 unverdicted novelty 6.0

    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.

  12. Diversity in Large Language Models under Supervised Fine-Tuning

    cs.LG 2026-04 unverdicted novelty 6.0

    TOFU loss mitigates the narrowing of generative diversity in LLMs after supervised fine-tuning by addressing neglect of low-frequency patterns and forgetting of prior knowledge.

  13. When to Ask a Question: Understanding Communication Strategies in Generative AI Tools

    cs.GT 2026-05 unverdicted novelty 5.0

    A tradeoff model shows generative AI can reduce bias against diverse preferences by strategically eliciting information instead of always inferring from majority patterns.

  14. Diversity in Large Language Models under Supervised Fine-Tuning

    cs.LG 2026-04 unverdicted novelty 5.0

    Supervised fine-tuning narrows LLM generative diversity through neglect of low-frequency patterns and knowledge forgetting, but the TOFU loss mitigates this effect across models and benchmarks.

  15. Polychromic Objectives for Reinforcement Learning

    cs.LG 2025-09 unverdicted novelty 5.0

    Introduces polychromic objectives adapted into PPO via vine sampling and modified advantages, showing higher success rates and better coverage under perturbations on BabyAI, Minigrid, and algorithmic tasks.

  16. AI as a Tool for Simulation-Based Experiments in Literary Studies

    cs.CL 2026-06 unverdicted novelty 4.0

    Proposes AI-driven simulations for literary-historical experiments and reports preliminary text-generation results claiming the first limited in-distribution outputs matching human novels.