A Survey of LLM Alignment: Instruction Understanding, Intention Reasoning, and Reliable Generation

TL;DR

This paper classify and analyze the performance of LLMs in challenging scenarios and conduct a comprehensive evaluation of existing solutions based on the aforementioned three core challenges: instruction understanding, intention reasoning, and reliable dialog generation.

Abstract

Large language models have demonstrated exceptional capabilities in understanding and generation. However, in real-world scenarios, users' natural language expressions are often inherently fuzzy, ambiguous, and uncertain, leading to challenges such as vagueness, polysemy, and contextual ambiguity. This paper focuses on three challenges in LLM-based text generation tasks: instruction understanding, intention reasoning, and reliable dialog generation. Regarding human complex instruction, LLMs have deficiencies in understanding long contexts and instructions in multi-round conversations. For intention reasoning, LLMs may have inconsistent command reasoning, difficulty reasoning about commands containing incorrect information, difficulty understanding user ambiguous language commands, and a weak understanding of user intention in commands. Besides, In terms of Reliable Dialog Generation, LLMs may have unstable generated content and unethical generation. To this end, we classify and analyze the performance of LLMs in challenging scenarios and conduct a comprehensive evaluation of existing solutions. Furthermore, we introduce benchmarks and categorize them based on the aforementioned three core challenges. Finally, we explore potential directions for future research to enhance the reliability and adaptability of LLMs in real-world applications.

Figures (11)

• Figure 1: Example of LLMs generation.
• Figure 2: Unlike previous surveys on LLMs, we do not consider the alignment between LLMs and humans as an isolated process,instead, we view it as a continuous and dynamic information processing process consisting of instruction understanding, intention reasoning, and reliable dialogue generation.
• Figure 3: Challenges and existing solutions between LLMs and Human Intentions.
• Figure 4: Case of Remote Information Failure (§\ref{['sec:long']}), where the model forgets relevant information over long distances in long context.
• Figure 5: Case of Incorrect Relevance Judgment(§\ref{['sec:multi']}), such as the model incorrectly associates wrong content from the previous turn.