Papers with commonsense reasoning and

3 papers
Contrastive Perplexity for Controlled Generation: An Application in Detoxifying Large Language Models (2025.acl-long)

Copied to clipboard

Challenge: Existing approaches to generate toxic content by large language models are based on pipelines . current approaches focus on preserving performance while effectively mitigating toxicity .
Approach: They propose a framework for implicit knowledge editing and controlled text generation by using hard negatives.
Outcome: The proposed framework significantly reduces toxic generation while maintaining strong performance on downstream tasks.
Task-Level Thinking Steps Help Large Language Models for Challenging Classification Task (2023.emnlp-main)

Copied to clipboard

Challenge: Experimental results prove the superiority of our proposed method on challenging classification tasks.
Approach: They propose a task-level thinking step that eliminates bias introduced by demonstrations . they propose 'progressive revision framework' which can improve the thinking steps by correcting hard demonstrations.
Outcome: The proposed method achieves best performance on three kinds of classification tasks in zero-shot and few-shot settings.
Empowering Multi-Turn Tool-Integrated Agentic Reasoning with Group Turn Policy Optimization (2026.acl-long)

Copied to clipboard

Challenge: Current reinforcement learning methods suffer from coarse-grained, trajectory-level rewards that provide insufficient learning signals for complex multi-turn interactions, leading to training stagnation.
Approach: They propose a novel RL algorithm for training large language models for multi-turn tool-integrated reasoning (TIR) that incorporates three innovations: turn-level reward assignment that provides fine-grained feedback for individual turns, return-based advantage estimation where normalized discounted returns are calculated as advantages, and self-supervised reward shaping that exploits self-supervision signals from generated code to densify sparse binary outcome-based rewards.
Outcome: The proposed algorithm outperforms GRPO by 3.0% across diverse math reasoning benchmarks and improves grepo by 3.9% on commonsense reasoning and program synthesis tasks.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations