Context Degradation Patterns

Context Degradation Patterns

Predictable degradation as context grows. Not binary - a continuum.

Degradation Patterns

Lost-in-Middle Phenomenon

• Information in middle gets 10-40% lower recall
• Models allocate massive attention to first token (BOS sink)
• As context grows, middle tokens fail to get sufficient attention
• Mitigation: Place critical info at beginning/end

[CURRENT TASK]              # Beginning - high attention
- Critical requirements

[DETAILED CONTEXT]          # Middle - lower attention
- Supporting details

[KEY FINDINGS]              # End - high attention
- Important conclusions

Context Poisoning

Entry points:

1. Tool outputs with errors/unexpected formats
2. Retrieved docs with incorrect/outdated info
3. Model-generated summaries with hallucinations

Detection symptoms:

• Degraded quality on previously successful tasks
• Tool misalignment (wrong tools/parameters)
• Persistent hallucinations

Recovery:

• Truncate to before poisoning point
• Explicit note + re-evaluation request
• Restart with clean context, preserve only verified info

Model Degradation Thresholds

Four-Bucket Mitigation

1. Write: Save externally (scratchpads, files)
2. Select: Pull only relevant (retrieval, filtering)
3. Compress: Reduce tokens (summarization)
4. Isolate: Split across sub-agents (partitioning)

Detection Heuristics

def calculate_health(utilization, degradation_risk, poisoning_risk):
    """Health score: 1.0 = healthy, 0.0 = critical"""
    score = 1.0
    score -= utilization * 0.5 if utilization > 0.7 else 0
    score -= degradation_risk * 0.3
    score -= poisoning_risk * 0.2
    return max(0, score)

# Thresholds: healthy >0.8, warning >0.6, degraded >0.4, critical <=0.4

Guidelines

1. Monitor context length vs performance correlation
2. Place critical info at beginning/end
3. Implement compaction before degradation
4. Validate retrieved docs before adding
5. Use versioning to prevent outdated clash
6. Segment tasks to prevent confusion
7. Design for graceful degradation

Related Topics

• Context Optimization - Mitigation techniques
• Multi-Agent Patterns - Isolation strategies