L-layer transformers under Log-ICoT curriculum provably learn k-parity with poly(n) samples and log k stages, matching explicit CoT efficiency without inference overhead.
super hub Mixed citations
Least-to-Most Prompting Enables Complex Reasoning in Large Language Models
Mixed citation behavior. Most common role is background (60%).
103
Pith papers citing it
Background
60% of classified citations
abstract
Chain-of-thought prompting has demonstrated remarkable performance on various natural language reasoning tasks. However, it tends to perform poorly on tasks which requires solving problems harder than the exemplars shown in the prompts. To overcome this challenge of easy-to-hard generalization, we propose a novel prompting strategy, least-to-most prompting. The key idea in this strategy is to break down a complex problem into a series of simpler subproblems and then solve them in sequence. Solving each subproblem is facilitated by the answers to previously solved subproblems. Our experimental results on tasks related to symbolic manipulation, compositional generalization, and math reasoning reveal that least-to-most prompting is capable of generalizing to more difficult problems than those seen in the prompts. A notable finding is that when the GPT-3 code-davinci-002 model is used with least-to-most prompting, it can solve the compositional generalization benchmark SCAN in any split (including length split) with an accuracy of at least 99% using just 14 exemplars, compared to only 16% accuracy with chain-of-thought prompting. This is particularly noteworthy because neural-symbolic models in the literature that specialize in solving SCAN are trained on the entire training set containing over 15,000 examples. We have included prompts for all the tasks in the Appendix.
hub tools
citation-role summary
background 14
method 4
dataset 1
other 1
citation-polarity summary
claims ledger
- abstract Chain-of-thought prompting has demonstrated remarkable performance on various natural language reasoning tasks. However, it tends to perform poorly on tasks which requires solving problems harder than the exemplars shown in the prompts. To overcome this challenge of easy-to-hard generalization, we propose a novel prompting strategy, least-to-most prompting. The key idea in this strategy is to break down a complex problem into a series of simpler subproblems and then solve them in sequence. Solving each subproblem is facilitated by the answers to previously solved subproblems. Our experimental
authors
co-cited works
representative citing papers
Promptbreeder evolves both task prompts and the mutation prompts that improve them using LLMs, outperforming Chain-of-Thought and Plan-and-Solve on arithmetic and commonsense reasoning benchmarks.
Tree of Thoughts enables language models to solve complex planning tasks by generating, evaluating, and searching over coherent intermediate thoughts in a tree, raising Game of 24 success from 4% to 74% with GPT-4.
PAL improves few-shot reasoning accuracy by having LLMs generate executable programs rather than text-based chains of thought, outperforming much larger models on math and logic benchmarks.
Language models generate robot policy code from natural language commands via few-shot prompting, enabling spatial-geometric reasoning, generalization, and precise control on real robots.
Rosetta Memory trains two profile-conditioned operators with a minimum-gain sampling curriculum and performance-gap reward to enable memory transfer between LLMs, showing gains on multi-hop QA benchmarks and robustness to unseen models.
Prefix gain measured via student-model solve-rate improvement is used to train a Prefix Utility Model (PUM) that supplies stronger supervision than correctness-based process rewards for mathematical reasoning.
An automatic numeric-remapping attack generator reveals 12-26 point accuracy drops on GSM8K for three LLMs while MAWPS and MultiArith stay near 98%.
Adding explicit parent pointers to represent search tree structure in LLM reasoning traces (LinTree) improves task performance and search efficiency on Blocks World, grid Navigation, and Sokoban relative to implicit traces and LLM-heuristic search.
ContextEcho benchmark shows persona drift occurs across 23 frontier models in long agentic-coding sessions, is not reliably reset by compaction, and can be restored by single-shot anchors with mode-dependent effects.
LGMT applies metamorphic testing derived from first-order logic equivalences to detect reasoning inconsistencies in LLMs that static benchmarks miss.
LEAD uses online adaptive mechanisms including Potential-Scaled Instability and symmetric efficiency rewards based on correct rollouts to achieve higher accuracy-efficiency scores with substantially shorter reasoning outputs than base models on math benchmarks.
Think-with-Rubrics has LLMs generate rubrics internally before responding, outperforming external rubric-as-reward baselines by 3.87 points on average across benchmarks.
Retrieving structured thinking traces as a corpus improves reasoning performance on AIME, LiveCodeBench, and GPQA over standard RAG or no retrieval.
VIDA provides 2,500 visually-dependent ambiguous translation examples and span-level disambiguation metrics; CoT-SFT on LVLMs improves out-of-distribution performance over standard SFT.
Presents MBFC-2025 dataset and multi-view embeddings with fusion methods for media bias and factuality, reporting SOTA results on ACL-2020 and new benchmarks on MBFC-2025.
Incisor uses program analysis and frontier LLMs to select working AWS EC2 instances ex ante for 100% of first-time HPC runs of C/C++/Fortran and Python codes, cutting runtime 54% and costs 44% versus an expert-constrained SkyPilot baseline.
Applying Canonical Correlation Analysis to paired residual activations from natural-language and symbolic reasoning chains in LLMs reveals a low-dimensional shared logical subspace that can steer the model's reasoning for up to 11 percentage point accuracy gains on logical benchmarks.
Self-Correcting RAG formalizes retrieval as MMKP to maximize information density under token limits and uses NLI-guided MCTS to validate faithfulness, raising accuracy and cutting hallucinations on six multi-hop QA and fact-checking datasets.
iTAG generates natural text paired with accurate causal graph annotations by framing concept assignment as an inverse problem and refining selections via chain-of-thought reasoning until the text's relations align with the target causal structure.
Chain-of-Thought prompting balances high accuracy with low energy use in small language models for code generation, while multi-sampling strategies add high energy costs for small accuracy gains.
LEAD lets LLMs solve checkers jumping puzzles up to size 13 by using lookahead to recover from irreversible errors on hard steps that break extreme decomposition.
Video-R1 uses temporal-aware RL and mixed datasets to boost video reasoning in MLLMs, with a 7B model reaching 37.1% on VSI-Bench and surpassing GPT-4o.
Coconut lets LLMs perform reasoning directly in continuous latent space by recycling hidden states as inputs, outperforming standard chain-of-thought on search-intensive logical tasks with better accuracy-efficiency trade-offs.
citing papers explorer
-
Transformers Provably Learn to Internalize Chain-of-Thought
L-layer transformers under Log-ICoT curriculum provably learn k-parity with poly(n) samples and log k stages, matching explicit CoT efficiency without inference overhead.
-
Promptbreeder: Self-Referential Self-Improvement Via Prompt Evolution
Promptbreeder evolves both task prompts and the mutation prompts that improve them using LLMs, outperforming Chain-of-Thought and Plan-and-Solve on arithmetic and commonsense reasoning benchmarks.
-
Tree of Thoughts: Deliberate Problem Solving with Large Language Models
Tree of Thoughts enables language models to solve complex planning tasks by generating, evaluating, and searching over coherent intermediate thoughts in a tree, raising Game of 24 success from 4% to 74% with GPT-4.
-
PAL: Program-aided Language Models
PAL improves few-shot reasoning accuracy by having LLMs generate executable programs rather than text-based chains of thought, outperforming much larger models on math and logic benchmarks.
-
Code as Policies: Language Model Programs for Embodied Control
Language models generate robot policy code from natural language commands via few-shot prompting, enabling spatial-geometric reasoning, generalization, and precise control on real robots.
-
Rosetta Memory: Adaptive Memory for Cross-LLM Agents
Rosetta Memory trains two profile-conditioned operators with a minimum-gain sampling curriculum and performance-gap reward to enable memory transfer between LLMs, showing gains on multi-hop QA benchmarks and robustness to unseen models.
-
From Correctness to Utility: Gain-Based Prefix Evaluation for LLM Reasoning
Prefix gain measured via student-model solve-rate improvement is used to train a Prefix Utility Model (PUM) that supplies stronger supervision than correctness-based process rewards for mathematical reasoning.
-
Testing LLM Arithmetic Reasoning Generalization with Automatic Numeric-Remapping Attacks
An automatic numeric-remapping attack generator reveals 12-26 point accuracy drops on GSM8K for three LLMs while MAWPS and MultiArith stay near 98%.
-
LinTree: Improving LLM Reasoning with Explicitly Structured Search Histories
Adding explicit parent pointers to represent search tree structure in LLM reasoning traces (LinTree) improves task performance and search efficiency on Blocks World, grid Navigation, and Sokoban relative to implicit traces and LLM-heuristic search.
-
ContextEcho: A Benchmark for Persona Drift in Long Agentic-Coding Sessions
ContextEcho benchmark shows persona drift occurs across 23 frontier models in long agentic-coding sessions, is not reliably reset by compaction, and can be restored by single-shot anchors with mode-dependent effects.
-
LGMT: Logic-Grounded Metamorphic Testing for Evaluating the Reasoning Reliability of LLMs
LGMT applies metamorphic testing derived from first-order logic equivalences to detect reasoning inconsistencies in LLMs that static benchmarks miss.
-
LEAD: Length-Efficient Adaptive and Dynamic Reasoning for Large Language Models
LEAD uses online adaptive mechanisms including Potential-Scaled Instability and symmetric efficiency rewards based on correct rollouts to achieve higher accuracy-efficiency scores with substantially shorter reasoning outputs than base models on math benchmarks.
-
Think-with-Rubrics: From External Evaluator to Internal Reasoning Guidance
Think-with-Rubrics has LLMs generate rubrics internally before responding, outperforming external rubric-as-reward baselines by 3.87 points on average across benchmarks.
-
RAG over Thinking Traces Can Improve Reasoning Tasks
Retrieving structured thinking traces as a corpus improves reasoning performance on AIME, LiveCodeBench, and GPQA over standard RAG or no retrieval.
-
VIDA: A dataset for Visually Dependent Ambiguity in Multimodal Machine Translation
VIDA provides 2,500 visually-dependent ambiguous translation examples and span-level disambiguation metrics; CoT-SFT on LVLMs improves out-of-distribution performance over standard SFT.
-
A Multi-View Media Profiling Suite: Resources, Evaluation, and Analysis
Presents MBFC-2025 dataset and multi-view embeddings with fusion methods for media bias and factuality, reporting SOTA results on ACL-2020 and new benchmarks on MBFC-2025.
-
Incisor: Ex Ante Cloud Instance Selection for HPC Jobs
Incisor uses program analysis and frontier LLMs to select working AWS EC2 instances ex ante for 100% of first-time HPC runs of C/C++/Fortran and Python codes, cutting runtime 54% and costs 44% versus an expert-constrained SkyPilot baseline.
-
Discovering a Shared Logical Subspace: Steering LLM Logical Reasoning via Alignment of Natural-Language and Symbolic Views
Applying Canonical Correlation Analysis to paired residual activations from natural-language and symbolic reasoning chains in LLMs reveals a low-dimensional shared logical subspace that can steer the model's reasoning for up to 11 percentage point accuracy gains on logical benchmarks.
-
Self-Correcting RAG: Enhancing Faithfulness via MMKP Context Selection and NLI-Guided MCTS
Self-Correcting RAG formalizes retrieval as MMKP to maximize information density under token limits and uses NLI-guided MCTS to validate faithfulness, raising accuracy and cutting hallucinations on six multi-hop QA and fact-checking datasets.
-
iTAG: Inverse Design for Natural Text Generation with Accurate Causal Graph Annotations
iTAG generates natural text paired with accurate causal graph annotations by framing concept assignment as an inverse problem and refining selections via chain-of-thought reasoning until the text's relations align with the target causal structure.
-
Evaluating the Environmental Impact of using SLMs and Prompt Engineering for Code Generation
Chain-of-Thought prompting balances high accuracy with low energy use in small language models for code generation, while multi-sampling strategies add high energy costs for small accuracy gains.
-
LEAD: Breaking the No-Recovery Bottleneck in Long-Horizon Reasoning
LEAD lets LLMs solve checkers jumping puzzles up to size 13 by using lookahead to recover from irreversible errors on hard steps that break extreme decomposition.
-
Video-R1: Reinforcing Video Reasoning in MLLMs
Video-R1 uses temporal-aware RL and mixed datasets to boost video reasoning in MLLMs, with a 7B model reaching 37.1% on VSI-Bench and surpassing GPT-4o.
-
Training Large Language Models to Reason in a Continuous Latent Space
Coconut lets LLMs perform reasoning directly in continuous latent space by recycling hidden states as inputs, outperforming standard chain-of-thought on search-intensive logical tasks with better accuracy-efficiency trade-offs.
-
Medusa: Simple LLM Inference Acceleration Framework with Multiple Decoding Heads
Medusa augments LLMs with multiple decoding heads and tree-based attention to predict and verify several tokens in parallel, yielding 2.2-3.6x inference speedup via two fine-tuning regimes.
-
Large Language Models as Optimizers
Large language models can optimize by being prompted with histories of past solutions and scores to propose better ones, producing prompts that raise accuracy up to 8% on GSM8K and 50% on Big-Bench Hard over human-designed baselines.
-
Visual ChatGPT: Talking, Drawing and Editing with Visual Foundation Models
Visual ChatGPT integrates visual foundation models with ChatGPT via prompts to enable multi-step image understanding, generation, and editing in conversational interactions.
-
A Prompt Pattern Catalog to Enhance Prompt Engineering with ChatGPT
The authors present a catalog of prompt patterns that provide reusable solutions to common problems in generating and interacting with outputs from LLMs.
-
Program of Thoughts Prompting: Disentangling Computation from Reasoning for Numerical Reasoning Tasks
PoT prompting improves numerical reasoning by having language models write programs executed by a computer instead of performing calculations in natural language chains of thought, with an average 12% gain over CoT.
-
VTI-CoT: Visual-Textual Interleaved Chain of Thought for Video Reasoning
VTI-CoT proposes a visual-textual interleaved chain-of-thought method for video reasoning, built via automated annotation and OCR compression, claiming SOTA performance and better training efficiency on same-scale models.
-
Latent Reward Steering: An Adaptive Inference-Time Framework that Implicitly Promotes Cognitive Behaviors in Reasoning LLMs
LRS trains a latent reward model on final-answer correctness to steer SAE states during inference, improving reasoning performance and implicitly encouraging better cognitive behaviors.
-
Efficient Table QA via TableGrid Navigation and Progressive Inference Prompting
Introduces TableGrid Navigation (TGN) and Progressive Inference Prompting (PIP) as training-free structured prompting frameworks that improve LLM performance on table question answering over baselines on TableBench and achieve SOTA on FeTaQa.
-
MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks
MT-JailBench is a modular benchmark that standardizes evaluation of multi-turn jailbreaks to identify key success drivers and enable stronger combined attacks.
-
PaT: Planning-after-Trial for Efficient Test-Time Code Generation
PaT defers planning until after failed trials in LLM code generation, enabling heterogeneous cheap-plus-powerful model setups that match large-model performance at roughly 69% lower cost.
-
Experience Sharing in Mutual Reinforcement Learning for Heterogeneous Language Models
Mutual Reinforcement Learning allows heterogeneous LLMs to exchange experience through mechanisms like Peer Rollout Pooling, Cross-Policy GRPO Advantage Sharing, and Success-Gated Transfer, with outcome-level sharing identified as favorable on the stability-support trade-off.
-
Learning Agent Routing From Early Experience
BoundaryRouter routes queries to LLM or agent using early experience memory from a seed set, cutting inference time 60.6% versus always using agents and raising performance 28.6% versus always using direct LLM inference.
-
Can RL Teach Long-Horizon Reasoning to LLMs? Expressiveness Is Key
RL training compute for logical reasoning follows a power law with horizon depth whose exponent rises with logical expressiveness, yielding better downstream transfer when models train on richer logics.
-
ReFlect: An Effective Harness System for Complex Long-Horizon LLM Reasoning
ReFlect is a harness that wraps LLMs to detect and recover from reasoning errors, achieving 7-29 pp gains over direct CoT on long-horizon tasks and improving code patch quality to 82-87%.
-
HypEHR: Hyperbolic Modeling of Electronic Health Records for Efficient Question Answering
HypEHR is a hyperbolic embedding model for EHR data that uses Lorentzian geometry and hierarchy-aware pretraining to answer clinical questions nearly as well as large language models but with much smaller size.
-
C-voting: Confidence-Based Test-Time Voting without Explicit Energy Functions
C-voting improves recurrent reasoning models by selecting among multiple latent trajectories the one with highest average top-1 probability, achieving 4.9% better Sudoku-hard accuracy than energy-based voting and outperforming HRM on Sudoku-extreme and Maze when paired with the new ItrSA++ model.
-
Structured Safety Auditing for Balancing Code Correctness and Content Safety in LLM-Generated Code
Dual Reasoning with explicit safety audits improves the new SUDS metric by 1.32x to 3.42x over baselines on code generation benchmarks containing injected harmful keywords.
-
When Verification Fails: How Compositionally Infeasible Claims Escape Rejection
AI claim verification models rely on salient-constraint shortcuts instead of full compositional reasoning under the closed-world assumption, as revealed by their over-acceptance of claims with supported salient constraints but contradicted non-salient ones.
-
FACT-E: Causality-Inspired Evaluation for Trustworthy Chain-of-Thought Reasoning
FACT-E uses controlled perturbations as an instrumental signal to measure intra-chain faithfulness in CoT reasoning and combines it with answer consistency to select trustworthy trajectories.
-
From Perception to Planning: Evolving Ego-Centric Task-Oriented Spatiotemporal Reasoning via Curriculum Learning
EgoTSR applies a three-stage curriculum on a 46-million-sample dataset to build egocentric spatiotemporal reasoning, reaching 92.4% accuracy on long-horizon tasks and reducing chronological biases.
-
Reason Only When Needed: Efficient Generative Reward Modeling via Model-Internal Uncertainty
E-GRM triggers CoT reasoning in generative reward models only when parallel generations show high uncertainty, reducing inference cost and raising accuracy on reasoning benchmarks via a hybrid regression-ranking scorer.
-
Generative AI Agent Empowered Power Allocation for HAP Propulsion and Communication Systems
A generative AI agent creates a realistic HAP propulsion power model including aerodynamic interference and enables a Q3E beamforming algorithm that improves QoS and energy efficiency.
-
ExecTune: Effective Steering of Black-Box LLMs with Guide Models
ExecTune trains guide models via acceptance sampling, supervised fine-tuning, and structure-aware RL to boost executability of strategies for black-box LLMs, yielding up to 9.2% higher accuracy and 22.4% lower cost on math and code tasks.
-
SeLaR: Selective Latent Reasoning in Large Language Models
SeLaR selectively applies latent soft reasoning in LLMs via entropy gating and contrastive regularization, outperforming standard CoT on five benchmarks without training.
-
Measuring Representation Robustness in Large Language Models for Geometry
LLMs display accuracy gaps of up to 14 percentage points on the same geometry problems solely due to representation choice, with vector forms consistently weakest and a convert-then-solve prompt helping only high-capacity models.
-
Sustainability Analysis of Prompt Strategies for SLM-based Automated Test Generation
Prompt strategies for SLM-based automated test generation vary widely in energy consumption and carbon emissions, with simpler strategies delivering competitive coverage at markedly lower environmental cost.