REVIEW 20 cited by
LLM Post-Training: A Deep Dive into Reasoning Large Language Models
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
read the original abstract
Large Language Models (LLMs) have transformed the natural language processing landscape and brought to life diverse applications. Pretraining on vast web-scale data has laid the foundation for these models, yet the research community is now increasingly shifting focus toward post-training techniques to achieve further breakthroughs. While pretraining provides a broad linguistic foundation, post-training methods enable LLMs to refine their knowledge, improve reasoning, enhance factual accuracy, and align more effectively with user intents and ethical considerations. Fine-tuning, reinforcement learning, and test-time scaling have emerged as critical strategies for optimizing LLMs performance, ensuring robustness, and improving adaptability across various real-world tasks. This survey provides a systematic exploration of post-training methodologies, analyzing their role in refining LLMs beyond pretraining, addressing key challenges such as catastrophic forgetting, reward hacking, and inference-time trade-offs. We highlight emerging directions in model alignment, scalable adaptation, and inference-time reasoning, and outline future research directions. We also provide a public repository to continually track developments in this fast-evolving field: https://github.com/mbzuai-oryx/Awesome-LLM-Post-training.
Forward citations
Cited by 20 Pith papers
-
S-GRPO: Unified Post-Training for Large Vision-Language Models
S-GRPO unifies SFT and RL for LVLMs via conditional ground-truth injection that supplies a maximal-reward anchor when group exploration fails completely.
-
TopoTuner: Topological Finetuning of Large Language Models
TopoTuner uses Wasserstein distances between persistence diagrams of attention projection weights to build reusable freezing profiles that match or beat LoRA while updating ~1-3% of model parameters.
-
AnySlot: Goal-Conditioned Vision-Language-Action Policies for Zero-Shot Slot-Level Placement
AnySlot decouples language slot selection from control via an explicit visual goal marker and reports large zero-shot gains on SlotBench, but the provided body text does not match the abstract.
-
Asking LLMs to Verify First is Almost Free Lunch
Prompting an LLM to first verify a random/trivial candidate answer before generating its solution consistently improves accuracy over chain-of-thought prompting at small extra token cost.
-
BigCharts-R1: Enhanced Chart Reasoning with Visual Reinforcement Finetuning
Replotting real-world charts into code-backed images, then fine-tuning with supervised learning and GRPO reinforcement learning, produces chart QA models that beat prior chart-specific models on several benchmarks.
-
Post-Training Large Language Models via Reinforcement Learning from Self-Feedback
RLSF uses a model's own answer-span confidence as an intrinsic reward to create preference data, then applies DPO or PPO to improve calibration and reasoning without external labels.
-
Re:Form -- Reducing Human Annotations in Scalable Formal Software Verification with RL in LLMs: A Preliminary Study on Dafny
A minimal-prior pipeline with automated data curation and verifier-driven RL lets small LLMs generate verifiable Dafny specifications and beat larger proprietary models on a synthetic compositional benchmark.
-
Draw an Ugly Person An Exploration of Generative AIs Perceptions of Ugliness
Generative image models disproportionately portray ugliness as old, white, and male, while verbally disclaiming the aesthetic judgments behind those images.
-
AdapThink: Adaptive Thinking Preferences for Reasoning Language Model
AdapThink is an RL post-training framework that adaptively reduces overthinking and underthinking in reasoning language models by rewarding confidence-appropriate reasoning depth and diverse training samples.
-
Spectral Origins of the Self-Correction Blind Spot in Autoregressive Generation
Self-correction blind spots in residual-stream autoregressive models arise iff the product of attention Jacobians has spectral radius ≥1, with a sharp marker threshold and RL coupling condition derived from that radius.
-
Inference-Time Policy Alignment for Fair Reinforcement Learning
A frozen RL policy can be reweighted at test time by a learned generalized-Gini welfare critic to improve fairness metrics, though the central equivalence mixes up two different welfare objectives.
-
RadReason: Radiology Report Evaluation Metric with Reasons and Sub-Scores
RadReason trains a 7B language model with GRPO to output six radiology error sub-scores plus textual reasons, reporting Kendall tau 0.730 on ReXVal, best among offline metrics.
-
Large Language Models Reasoning Abilities Under Non-Ideal Conditions After RL-Fine-Tuning
RL fine-tuning of LLMs improves clean-benchmark accuracy while degrading accuracy under three injected-distractor evaluation scenarios, though one of the three scenarios contradicts the headline claim.
-
Reinforcement Fine-Tuning Naturally Mitigates Forgetting in Continual Post-Training
Reinforcement fine-tuning largely prevents catastrophic forgetting during continual post-training of a multimodal LLM, while supervised fine-tuning degrades both task and general performance.
-
Graphs Meet AI Agents: Taxonomy, Progress, and Future Opportunities
A survey that groups graph-empowered AI agent research into planning, execution, memory, and multi-agent coordination, plus agents-for-graphs and applications.
-
From Meta-Thought to Execution: Cognitively Aligned Post-Training for Generalizable and Reliable LLM Reasoning
Post-training LLMs first on abstract number-free reasoning plans (CoMT), then with confidence-weighted rewards (CCRL), raises math accuracy by ~2-5 points over standard CoT-SFT+RL across four models.
-
Privacy-Preserving Approximate Nearest Neighbor Search on High-Dimensional Data
The submission's title and abstract describe a single-server privacy-preserving ANNS system with distance comparison encryption, but the manuscript body is an unrelated activity-recognition paper, leaving the advertis...
-
Large Language Model Powered Intelligent Urban Agents: Concepts, Capabilities, and Applications
The paper defines urban LLM agents, surveys their sensing, memory, reasoning, execution, and learning workflows, and organizes their applications across planning, transportation, environment, safety, and society.
-
Reinforcement Learning Meets Large Language Models: A Survey of Advancements and Applications Across the LLM Lifecycle
A survey that maps reinforcement learning methods, datasets, benchmarks, and open-source tools across the full training lifecycle of large language models, focusing on verifiable-reward reasoning.
-
Thinking Beyond Tokens: From Brain-Inspired Intelligence to Cognitive Foundations for Artificial General Intelligence and its Societal Impact
A broad survey arguing that AGI requires modular, memory-augmented, embodied architectures rather than scaled-up token prediction, with a brief proposal to decompose intelligence into five components.
Discussion (0). Sign in to comment.