Shadowgraph
About
Sign in

junseokkim00/ReASC

[ACL 2026] Reliability-Aware Adaptive Self-Consistency for Efficient Sampling in LLM Reasoning

0

stars

1

commits

Jupyter Notebook

primary language

Apr 15, 2026

updated

README

ReASC: Reliability-Aware Adaptive Self-Consistency for Efficient Sampling in LLM Reasoning

The Official code for Reliability-Aware Adaptive Self-Consistency for Efficient Sampling in LLM Reasoning

  1. generate conda environment
$ conda env create -f environment.yaml
$ conda activate reasc
  1. run run_self_certainty_calibration.sh to get result for calibration set in offline settings
  2. edit run_sc_self_certainty.sh and run it for the result of standard Self Consistency.
  3. run each notebook within ./notebooks for the result of ASC, ESC, ReASC Offline and ReASC Online

Contributors

junseokkim00

1 commits

Similar repos

jinhaoduan/SAR

[ACL 2024] Shifting Attention to Relevance: Towards the Predictive Uncertainty Quantification of…

64

chaeyoon-jang/calibrag

CalibRAG: Calibrated Decision-Making through Large LLM-Assisted Retrieval

4

xiaowu0162/sync-ralm-faithfulness

Synchronous faithfulness checking and decoding for RALMs

4

YecanLee/Adaptive-Contrastive-Search

[EMNLP 2024 Findings] Official PyTorch Implementation of "Adaptive Contrastive Search:…

42

TaoShuchang/CONQORD

The repository for ACL 2024 paper "When to Trust LLMs: Aligning Confidence with Response Quality"

8

wataruhashimoto52/decoding_uncertainty

Code for "Decoding Uncertainty: The Impact of Decoding Strategies for Uncertainty Estimation in…

1

MiaoXiong2320/llm-uncertainty

code repo for ICLR 2024 paper "Can LLMs Express Their Uncertainty? An Empirical Evaluation of…

149

learnsyslab/uq_vla

Official code repository for "Uncertainty Quantification for Flow-Based Vision-Language-Action…

22

Languages

Jupyter Notebook

98.0%

Python

1.9%

junseokkim00/ReASC

[ACL 2026] Reliability-Aware Adaptive Self-Consistency for Efficient Sampling in LLM Reasoning

0

stars

1

commits

Jupyter Notebook

primary language

Apr 15, 2026

updated

README

ReASC: Reliability-Aware Adaptive Self-Consistency for Efficient Sampling in LLM Reasoning

The Official code for Reliability-Aware Adaptive Self-Consistency for Efficient Sampling in LLM Reasoning

  1. generate conda environment
$ conda env create -f environment.yaml
$ conda activate reasc
  1. run run_self_certainty_calibration.sh to get result for calibration set in offline settings
  2. edit run_sc_self_certainty.sh and run it for the result of standard Self Consistency.
  3. run each notebook within ./notebooks for the result of ASC, ESC, ReASC Offline and ReASC Online

Similar repos

jinhaoduan/SAR

[ACL 2024] Shifting Attention to Relevance: Towards the Predictive Uncertainty Quantification of…

64

chaeyoon-jang/calibrag

CalibRAG: Calibrated Decision-Making through Large LLM-Assisted Retrieval

4

xiaowu0162/sync-ralm-faithfulness

Synchronous faithfulness checking and decoding for RALMs

4

YecanLee/Adaptive-Contrastive-Search

[EMNLP 2024 Findings] Official PyTorch Implementation of "Adaptive Contrastive Search:…

42

TaoShuchang/CONQORD

The repository for ACL 2024 paper "When to Trust LLMs: Aligning Confidence with Response Quality"

8

wataruhashimoto52/decoding_uncertainty

Code for "Decoding Uncertainty: The Impact of Decoding Strategies for Uncertainty Estimation in…

1

MiaoXiong2320/llm-uncertainty

code repo for ICLR 2024 paper "Can LLMs Express Their Uncertainty? An Empirical Evaluation of…

149

learnsyslab/uq_vla

Official code repository for "Uncertainty Quantification for Flow-Based Vision-Language-Action…

22

Contributors

junseokkim00

1 commits

Languages

Jupyter Notebook

98.0%

Python

1.9%