Insights into LLM Long-Context Failures: When Transformers Know but Don’t Tell (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) exhibit positional bias, struggling to utilize information from the middle or end of long contexts. |
| Approach: | They propose to examine LLMs' long-context generalizations by probing their hidden representations. |
| Outcome: | The proposed models excel at processing extended contexts while preserving their positional bias. |
Similar Papers
What Context Features Can Transformer Language Models Use? (2021.acl-long)
Copied to clipboard
| Challenge: | Recent studies show that transformer-based language models benefit from conditioning on contexts of hundreds to thousands of previous tokens. |
| Approach: | They propose to use lexical and structural information to ablate usable information in transformer language models. |
| Outcome: | The proposed model improves when conditioning on contexts of thousands of previous tokens. |
Self-Consistency Falls Short! The Adverse Effects of Positional Bias on Long-Context Problems (2026.tacl-1)
Copied to clipboard
| Challenge: | Existing approaches to improve long-context understanding of large language models lack scalability and stability. |
| Approach: | They challenge the assumption that SC’s benefits generalize to long-context settings . they find that persistent position bias degrades performance on long-consistency tasks . |
| Outcome: | The proposed approach fails to improve and actively degrades performance on long-context tasks. |
Language Models Struggle to Use Representations Learned In-Context (2026.acl-long)
Copied to clipboard
| Challenge: | a recent study shows that large language models are capable of inducing rich representations of data that are seen in-context . a novel task, adaptive world modeling, shows that even the most performant LLMs cannot reliably leverage novel semantics defined in-constitut. |
| Approach: | They propose to use in-context representations to induce rich representations of data . they also propose to probe models using a novel task to enable flexible deployment . |
| Outcome: | The proposed model can use in-context representations to complete simple downstream tasks. |
Can Calibration of Positional Encodings Enhance Long Context Utilization? (2026.findings-eacl)
Copied to clipboard
| Challenge: | Large language models suffer from positional biases that reduce effective utilization of long contexts. |
| Approach: | They propose a training-free framework for calibrating Positional Encodings at inference time. |
| Outcome: | The proposed framework improves on needle-in-a-haystack and cross-chunk reasoning benchmarks and provides a lightweight method for improving long-context utilization. |
Found in the middle: Calibrating Positional Attention Bias Improves Long Context Utilization (2024.findings-acl)
Copied to clipboard
Cheng-Yu Hsieh, Yung-Sung Chuang, Chun-Liang Li, Zifeng Wang, Long Le, Abhishek Kumar, James Glass, Alexander Ratner, Chen-Yu Lee, Ranjay Krishna, Tomas Pfister
| Challenge: | Large language models struggle to capture relevant information located in the middle of their input. |
| Approach: | They propose a calibration mechanism that allows the model to attend to contexts faithfully according to their relevance even when they are in the middle. |
| Outcome: | The proposed calibration mechanism mitigates this positional bias and improves retrieval-augmented generation performance. |
Efficient Solutions For An Intriguing Failure of LLMs: Long Context Window Does Not Mean LLMs Can Analyze Long Sequences Flawlessly (2025.coling-main)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have demonstrated remarkable capabilities in comprehending and analyzing lengthy sequential inputs. |
| Approach: | They propose to implement ad-hoc solutions that enhance LLMs’ performance on long input sequences by up to 50% while reducing API cost and latency by up . to address this limitation, they propose to use three datasets and two tasks to analyze news categorization and sentence analysis to evaluate their models. |
| Outcome: | The proposed solutions significantly improve LLMs’ performance on long input sequences by up to 50% while reducing API cost and latency by up . to 93% and 50%, respectively. |
Context Length Alone Hurts LLM Performance Despite Perfect Retrieval (2025.findings-emnlp)
Copied to clipboard
Yufeng Du, Minyang Tian, Srikanth Ronanki, Subendhu Rongali, Sravan Babu Bodapati, Aram Galstyan, Azton Wells, Roy Schwartz, Eliu A Huerta, Hao Peng
| Challenge: | Large language models (LLMs) often fail to scale their performance on long-context tasks performance in line with the context lengths they support. |
| Approach: | They propose a model-agnostic mitigation strategy that transforms a long-context task into a short-concept one by prompting the model to recite the retrieved evidence before attempting to solve the problem. |
| Outcome: | The proposed model improves on a long-context task up to 4% on RULER. |
Lost in the Distance: Large Language Models Struggle to Capture Long-Distance Relational Knowledge (2025.findings-naacl)
Copied to clipboard
| Challenge: | Recent large language models have demonstrated impressive capabilities in handling long contexts . however, as context length increases, LLMs struggle more with filtering out irrelevant information . |
| Approach: | They propose to use unrelated sentences to capture relational knowledge over long contexts . they find that LLMs can handle edge noise with little impact, but can reason about distant relationships . |
| Outcome: | The proposed model can handle edge noise with little impact, but its ability to reason about distant relationships declines as the noise grows. |
On Positional Bias of Faithfulness for Long-form Summarization (2025.naacl-long)
Copied to clipboard
| Challenge: | Large language models exhibit positional bias in long-context settings, under-attending to information in the middle. |
| Approach: | They compile eight human-annotated long-form summarization datasets to evaluate faithfulness . they find that LLMs faithfully summarize beginning and end of documents but neglect middle content . |
| Outcome: | The proposed methods show that LLMs under-attend to information in the middle of inputs. |
Racing Thoughts: Explaining Contextualization Errors in Large Language Models (2025.naacl-long)
Copied to clipboard
| Challenge: | Large Language Models have demonstrated a remarkable capacity for accomplishing a wide variety of language generation and classification tasks. |
| Approach: | They propose a race conditions hypothesis to explain contextualization errors . they propose to use a variety of techniques to test the hypothesis . |
| Outcome: | The proposed model fails to properly contextualize a financial institution if it does not include a bank . the proposed model is based on the race conditions hypothesis . |