DSPy compiles short declarative programs into LM pipelines that self-optimize and outperform both standard few-shot prompting and expert-written chains on math, retrieval, and QA tasks.
hub Canonical reference
Large Language Models are Zero-Shot Reasoners
Canonical reference. 88% of citing Pith papers cite this work as background.
86
Pith papers citing it
1,108
external citations · Pith
Background
88% of classified citations
abstract
Pretrained large language models (LLMs) are widely used in many sub-fields of natural language processing (NLP) and generally known as excellent few-shot learners with task-specific exemplars. Notably, chain of thought (CoT) prompting, a recent technique for eliciting complex multi-step reasoning through step-by-step answer examples, achieved the state-of-the-art performances in arithmetics and symbolic reasoning, difficult system-2 tasks that do not follow the standard scaling laws for LLMs. While these successes are often attributed to LLMs' ability for few-shot learning, we show that LLMs are decent zero-shot reasoners by simply adding "Let's think step by step" before each answer. Experimental results demonstrate that our Zero-shot-CoT, using the same single prompt template, significantly outperforms zero-shot LLM performances on diverse benchmark reasoning tasks including arithmetics (MultiArith, GSM8K, AQUA-RAT, SVAMP), symbolic reasoning (Last Letter, Coin Flip), and other logical reasoning tasks (Date Understanding, Tracking Shuffled Objects), without any hand-crafted few-shot examples, e.g. increasing the accuracy on MultiArith from 17.7% to 78.7% and GSM8K from 10.4% to 40.7% with large InstructGPT model (text-davinci-002), as well as similar magnitudes of improvements with another off-the-shelf large model, 540B parameter PaLM. The versatility of this single prompt across very diverse reasoning tasks hints at untapped and understudied fundamental zero-shot capabilities of LLMs, suggesting high-level, multi-task broad cognitive capabilities may be extracted by simple prompting. We hope our work not only serves as the minimal strongest zero-shot baseline for the challenging reasoning benchmarks, but also highlights the importance of carefully exploring and analyzing the enormous zero-shot knowledge hidden inside LLMs before crafting finetuning datasets or few-shot exemplars.
hub tools
citation-role summary
background 14
baseline 1
method 1
citation-polarity summary
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.
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.
An automatic numeric-remapping attack generator reveals 12-26 point accuracy drops on GSM8K for three LLMs while MAWPS and MultiArith stay near 98%.
AuDisAgent reformulates multimodal controversy detection as a dynamic audience dissemination process using screening, panel discussion, and arbitration agents, plus comment bootstrapping, and reports outperforming prior static methods on a public dataset.
Banning filler words like 'very' and 'just' improved LLM reasoning by 6.7 percentage points while E-Prime improved it by only 3.7, with gains ranking in exact inverse order of theoretical depth across models and tasks.
Reasoning LLMs with minimal tools for tree construction and analysis induce decision trees that outperform CART, compete with ensembles on low-resource tabular data, and provide human-readable reasoning traces.
LLMs display high variance and major accuracy drops on GSM-Symbolic variants of grade-school math problems, indicating they replicate training patterns rather than execute logical reasoning.
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.
VoxPoser uses LLMs to compose 3D value maps via VLM interaction for model-based synthesis of robust robot trajectories on open-set language-specified manipulation tasks.
Process supervision significantly outperforms outcome supervision for training models on the MATH dataset, achieving 78% accuracy on a representative test subset with active learning and a released 800k step-label dataset.
GPT-4 exceeds the USMLE passing score by more than 20 points and outperforms both GPT-3.5 and the medically fine-tuned Med-PaLM on the MultiMedQA benchmarks.
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.
Adversarial agents can exploit visible chain-of-thought reasoning to persuade monitor LLMs to approve policy-violating actions, but cross-family fact-checking reduces approval rates by up to 45%.
Forced CoT produces video-dependent reasoning chains but does not improve MCQ accuracy on Qwen2.5-VL with Video-MME and causes a small drop on the 7B variant.
Guava harness enables 4B open-source models to achieve performance comparable to frontier models on embodied manipulation tasks by distilling capabilities from under 2K simulation trajectories using three identified design principles.
ReLAR uses reinforcement-guided latent refinement with adaptive controllers to improve LLM reasoning accuracy and stability at lower inference cost than explicit reasoning methods.
Dropout-GRPO uses structured dropout to generate trajectory variance for GRPO in latent-reasoning models like Coconut, raising GSM8K pass@1 from 27.29% to 29.01%.
R3LM trains LLMs via two-stage reasoning-then-regression on a new dataset CRE-ReasonBench with mechanistic traces, achieving SOTA enhancer activity prediction across three cell types with interpretable outputs.
LLMs flag byte-identical errors 23-93 points more often when the error is presented under an external chat role than inside their own <thought> block, so role labeling, not content, gates explicit self-correction.
A small GPT-2 model trained from scratch on GASING-derived CoT supervision for arithmetic reaches over 80% held-out accuracy, exhibits three learning phases, and develops both procedural and associative reasoning.
TRI trains LLMs on goal-conditioned fill-in-the-middle tasks via PSM token rearrangement and symbolic verification to surgically repair erroneous CoT segments.
TravelEval is a new benchmark with a six-dimensional evaluation framework, realistic data sandbox, and simulation-based global assessment for LLM-powered travel planning agents.
Reward-guided optimization of SAE latents, gated by reward and confidence, improves multi-benchmark reasoning and is post-hoc associated with more verification and course correction.
citing papers explorer
-
DSPy: Compiling Declarative Language Model Calls into Self-Improving Pipelines
DSPy compiles short declarative programs into LM pipelines that self-optimize and outperform both standard few-shot prompting and expert-written chains on math, retrieval, and QA tasks.
-
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.
-
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.
-
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%.
-
From Static Analysis to Audience Dissemination: A Training-Free Multimodal Controversy Detection Multi-Agent Framework
AuDisAgent reformulates multimodal controversy detection as a dynamic audience dissemination process using screening, panel discussion, and arbitration agents, plus comment bootstrapping, and reports outperforming prior static methods on a public dataset.
-
Trivial Vocabulary Bans Improve LLM Reasoning More Than Deep Linguistic Constraints
Banning filler words like 'very' and 'just' improved LLM reasoning by 6.7 percentage points while E-Prime improved it by only 3.7, with gains ranking in exact inverse order of theoretical depth across models and tasks.
-
Talking Trees: Reasoning-Assisted Induction of Decision Trees for Tabular Data
Reasoning LLMs with minimal tools for tree construction and analysis induce decision trees that outperform CART, compete with ensembles on low-resource tabular data, and provide human-readable reasoning traces.
-
GSM-Symbolic: Understanding the Limitations of Mathematical Reasoning in Large Language Models
LLMs display high variance and major accuracy drops on GSM-Symbolic variants of grade-school math problems, indicating they replicate training patterns rather than execute logical reasoning.
-
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.
-
VoxPoser: Composable 3D Value Maps for Robotic Manipulation with Language Models
VoxPoser uses LLMs to compose 3D value maps via VLM interaction for model-based synthesis of robust robot trajectories on open-set language-specified manipulation tasks.
-
Let's Verify Step by Step
Process supervision significantly outperforms outcome supervision for training models on the MATH dataset, achieving 78% accuracy on a representative test subset with active learning and a released 800k step-label dataset.
-
Capabilities of GPT-4 on Medical Challenge Problems
GPT-4 exceeds the USMLE passing score by more than 20 points and outperforms both GPT-3.5 and the medically fine-tuned Med-PaLM on the MultiMedQA benchmarks.
-
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.
-
Persuasion Attacks Can Decrease Effectiveness of CoT Monitoring
Adversarial agents can exploit visible chain-of-thought reasoning to persuade monitor LLMs to approve policy-violating actions, but cross-family fact-checking reduces approval rates by up to 45%.
-
Chains That See, Answers That Don't: A Multi-Aspect Evaluation Recipe for Forced Chain-of-Thought on Video-MME
Forced CoT produces video-dependent reasoning chains but does not improve MCQ accuracy on Qwen2.5-VL with Video-MME and causes a small drop on the 7B variant.
-
Guava: An Effective and Universal Harness for Embodied Manipulation
Guava harness enables 4B open-source models to achieve performance comparable to frontier models on embodied manipulation tasks by distilling capabilities from under 2K simulation trajectories using three identified design principles.
-
Learning to Refine Hidden States for Reliable LLM Reasoning
ReLAR uses reinforcement-guided latent refinement with adaptive controllers to improve LLM reasoning accuracy and stability at lower inference cost than explicit reasoning methods.
-
Dropout-GRPO: Variational Stochasticity for Continuous Latent Reasoning
Dropout-GRPO uses structured dropout to generate trajectory variance for GRPO in latent-reasoning models like Coconut, raising GSM8K pass@1 from 27.29% to 29.01%.
-
Biological Reasoning-Informed Regression for Interpretable Regulatory DNA Activity Prediction
R3LM trains LLMs via two-stage reasoning-then-regression on a new dataset CRE-ReasonBench with mechanistic traces, achieving SOTA enhancer activity prediction across three cell types with interpretable outputs.
-
The Self-Correction Illusion: Role Relabeling Gates Explicit Error Flagging in Large Language Models
LLMs flag byte-identical errors 23-93 points more often when the error is presented under an external chat role than inside their own <thought> block, so role labeling, not content, gates explicit self-correction.
-
Arithmetic Pedagogy for Language Models
A small GPT-2 model trained from scratch on GASING-derived CoT supervision for arithmetic reaches over 80% held-out accuracy, exhibits three learning phases, and develops both procedural and associative reasoning.
-
Imbuing Large Language Models with Bidirectional Logic for Robust Chain Repair
TRI trains LLMs on goal-conditioned fill-in-the-middle tasks via PSM token rearrangement and symbolic verification to surgically repair erroneous CoT segments.
-
TravelEval: A Comprehensive Benchmarking Framework for Evaluating LLM-Powered Travel Planning Agents
TravelEval is a new benchmark with a six-dimensional evaluation framework, realistic data sandbox, and simulation-based global assessment for LLM-powered travel planning agents.
-
Latent Reward Steering: An Adaptive Inference-Time Framework that Implicitly Promotes Cognitive Behaviors in Reasoning LLMs
Reward-guided optimization of SAE latents, gated by reward and confidence, improves multi-benchmark reasoning and is post-hoc associated with more verification and course correction.
-
The Sword, Shield, and Achilles' Heel: Characterizing the Linguistic Inductive Bias of Large Language Models for Spatial Reasoning in Navigation Planning
Experiments reveal that topological cues robustly support LLM navigation planning while incorrect semantic cues derail it, with linguistic format effects varying by model size and compression.
-
The Curse of Helpfulness: Inverse Scaling Law in Robustness to Distractor Instructions via DistractionIF
DistractionIF benchmark reveals inverse scaling in LLM robustness to distractors in reference text, with GRPO RL as a mitigation.
-
Strategy-Induct: Task-Level Strategy Induction for Instruction Generation
Strategy-Induct induces task-level instructions from question-only examples by generating reasoning strategies first, then using those pairs to create a guiding instruction.
-
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.
-
Memory Inception: Latent-Space KV Cache Manipulation for Steering LLMs
Memory Inception is a training-free method that injects latent KV banks at chosen layers to steer LLMs, achieving superior control-drift balance and up to 118x storage reduction on personality and structured-reasoning tasks.
-
Compared to What? Baselines and Metrics for Counterfactual Prompting
Counterfactual prompting effects on LLMs are often indistinguishable from those caused by meaning-preserving paraphrases, causing most previously reported demographic sensitivities to disappear under proper statistical comparison.
-
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.
-
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.
-
Measuring and curing reasoning rigidity: from decorative chain-of-thought to genuine faithfulness
SLRC quantifies genuine step necessity in LLM reasoning as a causal estimator, LC-CoSR training reduces rigidity with stability guarantees, and evaluations reveal a faithfulness-sycophancy paradox across frontier models.
-
GraphMind: Theorem Selection and Conclusion Generation Framework with Dynamic GNN for LLM Reasoning
GraphMind models multi-step reasoning as an evolving heterogeneous graph, using GNN encoding and semantic matching to select theorems and generate conclusions iteratively, reporting performance gains over baselines on QA datasets.
-
ARM: Discovering Agentic Reasoning Modules for Generalizable Multi-Agent Systems
ARM evolves specialized reasoning modules from basic CoT via tree search to serve as reusable components in multi-agent systems that generalize across models and domains without per-task re-optimization.
-
Towards an AI co-scientist
A multi-agent AI system generates novel biomedical hypotheses that show promising experimental validation in drug repurposing for leukemia, new targets for liver fibrosis, and a bacterial gene transfer mechanism.
-
Compressed Chain of Thought: Efficient Reasoning Through Dense Representations
CCoT generates variable-length continuous contemplation tokens that compress explicit reasoning chains, enabling additional dense reasoning and accuracy gains in off-the-shelf language models while allowing adaptive control of token count.
-
Inference Scaling Laws: An Empirical Analysis of Compute-Optimal Inference for Problem-Solving with Language Models
Empirical analysis shows scaling inference compute via strategies like tree search can be more efficient than scaling model parameters, with 7B models plus novel search outperforming 34B models.
-
Scaling Data-Constrained Language Models
Repeating training data up to 4 epochs yields negligible loss increase versus unique data for fixed compute, and a new scaling law accounts for the decaying value of repeated tokens and excess parameters.
-
Gorilla: Large Language Model Connected with Massive APIs
Gorilla is a fine-tuned LLM that surpasses GPT-4 in accurate API call generation and uses retrieval to handle documentation updates.
-
Reasoning with Language Model is Planning with World Model
RAP turns LLMs into dual world-model and planning agents via MCTS to generate better reasoning paths, outperforming CoT baselines and achieving 33% relative gains over GPT-4 CoT using LLaMA-33B on plan generation.
-
Improving Factuality and Reasoning in Language Models through Multiagent Debate
Multiagent debate among LLMs improves mathematical reasoning, strategic reasoning, and factual accuracy while reducing hallucinations.
-
Towards Expert-Level Medical Question Answering with Large Language Models
Med-PaLM 2 achieves 86.5% accuracy on MedQA and approaches or exceeds prior state-of-the-art on other medical QA benchmarks while receiving higher physician preference ratings than human answers on consumer questions.
-
Distilling Step-by-Step! Outperforming Larger Language Models with Less Training Data and Smaller Model Sizes
Distilling step-by-step uses LLM-generated rationales as additional supervision in a multi-task framework so that 770M-parameter models outperform 540B-parameter models on NLP benchmarks with only 80% of the data.
-
CAMEL: Communicative Agents for "Mind" Exploration of Large Language Model Society
CAMEL proposes a role-playing framework with inception prompting that enables autonomous multi-agent cooperation among LLMs and generates conversational data for studying their behaviors.
-
MM-REACT: Prompting ChatGPT for Multimodal Reasoning and Action
MM-REACT uses textual prompts to let ChatGPT collaborate with external vision experts for zero-shot multimodal reasoning and action on advanced visual tasks.
-
ART: Automatic multi-step reasoning and tool-use for large language models
ART automatically generates multi-step reasoning programs with tool integration for LLMs, yielding substantial gains over few-shot and auto-CoT prompting on BigBench and MMLU while matching hand-crafted CoT on most tasks.
-
PaLM-E: An Embodied Multimodal Language Model
PaLM-E is a single 562B-parameter multimodal model that performs embodied reasoning tasks like robotic manipulation planning and visual question answering by interleaving vision, state, and text inputs with positive transfer from joint training on language and robotics data.
-
Multimodal Chain-of-Thought Reasoning in Language Models
Multimodal-CoT achieves state-of-the-art on ScienceQA by using a two-stage process that incorporates vision into chain-of-thought rationale generation for models under 1 billion parameters.
-
Solving math word problems with process- and outcome-based feedback
On GSM8K, outcome-based supervision achieves similar final-answer error rates to process-based with less labeling, but process-based or learned reward models are needed to reach 3.4% reasoning error among correct solutions.