Papersby GlyphTruthRL: Incentivizing Truthful LLMs via Reinforcement Learning research paper by Meta, 2025
Meta · Foundation models · Sep 30, 2025 · 24 citations · 56 upvotes · unverified
What it shows
TruthRL, a reinforcement learning framework, enhances the truthfulness of large language models by balancing accuracy and abstention, significantly reducing hallucinations and improving performance across benchmarks.
By Zhepei Wei, Xiao Yang, Kai Sun and 12 more · arXiv 2509.25760 · PDF
UnverifiedHugging Face's summary; not yet checked by hand.
More from Meta
All 40InteractiveWorld: A Verifiable Human-Centric Environment for Building Better Interactive MLLM AssistantsInteractiveWorld provides a verifiable simulation environment and benchmark to evaluate whether multimodal language models can reliably guide humans through physical tasks.
Meta46 upvotesRead →Native Action-Prior Learning from Videos for World Action ModelsNAVA-WAM learns robot action priors directly from observation-only videos, improving control performance without needing action-labeled data for pretraining.Meta88 upvotesRead →Sharpening Tax in Post-TrainingThis paper introduces a metric to quantify the loss in solution coverage after reinforcement learning and proposes a sampler to mitigate it.Meta1 citations111 upvotesRead →World Action Modeling with Progressive Visual PlanningProWAM improves robotic control by predicting sparse visual sub-goals to guide action generation, avoiding the need for dense video generation.Meta1 citations98 upvotesRead →Context Language ModelsContext Language Models manage their own context by treating it as an editable file, improving performance and efficiency.Meta5 citations47 upvotesRead →In-Flight KV Cache with Clean Anchors for Faster Autoregressive Video DiffusionFlashForward accelerates video diffusion by reusing in-flight KV caches and using sparse clean anchors to maintain structural consistency.Meta1 citations34 upvotesRead →WearableQA: A Benchmark for Health Reasoning over Real-World Wearable DataWearableQA is a benchmark of multiple-choice questions derived from real longitudinal wearable data that evaluates large language model reasoning across data and health dimensions.Meta43 upvotesRead →Towards Physics of Multimodal Pretraining: Knowledge Flow, Modality Synergy, Early Unification, and RecipesControlled experiments on when text, image understanding and image generation help or compete when trained together.Meta7 citations63 upvotesRead →
Other foundation models papers
TopicBERT: Pre-training of Deep Bidirectional Transformers for Language UnderstandingPre-training a Transformer to fill in masked words, then fine-tuning it, beat task-specific models across language understanding tests.
Google125k citations33 upvotesRead →Language Models are Few-Shot LearnersGPT-3, a 175B-parameter model, does new tasks from a few examples in the prompt, with no fine-tuning.
OpenAI66.4k citations21 upvotesRead →GPT-4 Technical ReportA multimodal model that passes a simulated bar exam with a score around the top 10% of test takers.OpenAI28.4k citations7 upvotesRead →Qwen3 Technical ReportQwen3, a unified series of large language models, integrates thinking and non-thinking modes, reduces computational resources, and achieves state-of-the-art performance across various tasks and languages.
Alibaba (Qwen)10.1k citations339 upvotesRead →Gemini 1.5: Unlocking multimodal understanding across millions of tokens of contextRecalls details across millions of tokens of text, hours of video and days of audio.
Google DeepMind4,190 citations65 upvotesRead →Qwen3-VL Technical ReportQwen3-VL, a vision-language model, excels in text and multimodal understanding through advanced architectures and larger contexts, achieving superior performance across benchmarks.Alibaba (Qwen)3,178 citations164 upvotesRead →
About this paper
- Venue
- arXiv.org
- Citations
- 24, 5 influential · Semantic Scholar
- Upvotes
- 56 · Hugging Face
- Lab
- Meta · on Companies · on Acquisitions · on Quarterly · on Paydays · on TechConf
Changes
Influential citations rose from 3 to 5.
Citations rose from 20 to 24.
Citations rose from 18 to 20.
Influential citations rose from 2 to 3.
Citations rose from 17 to 18.
A first count of 2 influential citations.
A first count of 17 citations.
Found by the weekly Hugging Face scan; not yet checked by hand.