Effetto «lost in the middle»
I modelli leggono l’inizio e la fine in modo più affidabile rispetto alla parte centrale.
Effetto «lost in the middle» è una lezione Claude Architect gratuita su CoddyKit. Questa è la lezione 3 di 4. Puoi leggere la lezione completa qui gratuitamente — poi esercitati direttamente nel browser con un editor di codice integrato e un tutor IA disponibile 24/7. Fa parte del percorso di apprendimento Claude Architect, e i tuoi progressi si sincronizzano tra il web e l'app CoddyKit. Il corso Claude Architect include 4 lezioni in totale.
Parti di questa lezione non sono ancora state tradotte e vengono mostrate in inglese.
The Effect, Stated Plainly
When you pack a long prompt into Claude's context window, the model does not attend to every token equally. Information near the start and the end of the context is recalled far more reliably than information buried in the middle.
This is the lost-in-the-middle effect. It is not a bug you can patch — it is a property of how attention behaves over long inputs. As an architect, you design around it.
Why It Matters for Reliability
On the Claude Certified Architect exam, this sits in Domain 5: Context Management & Reliability. The practical risk: a critical instruction, policy rule, or transactional fact that you placed in the middle of a bloated prompt gets silently ignored.
The failure is quiet. The model still produces a fluent answer — it just dropped the constraint you cared about. That makes lost-in-the-middle a reliability problem, not a formatting preference.
Position Your Key Instructions
The simplest mitigation: put your most important instructions where attention is strongest. Lead with them in the system prompt, and restate the critical constraint at the end, right before the model generates.
Remember the API contract: the model keeps no state. You send the FULL messages history every turn. So every turn is a fresh chance — and a fresh risk — for middle content to be under-weighted.
resp = client.messages.create(
model="claude-sonnet-4-5",
max_tokens=1024,
system=(
"You are a refund agent. CRITICAL RULE: never "
"process a refund before get_customer returns a "
"verified ID."
),
# ...long retrieved context goes in the middle...
messages=messages,
)Bookend the Critical Constraint
A robust pattern is to bookend: state the rule up front in system, then echo a short reminder of it as the final line of the last user turn. The start anchors it; the end keeps it top-of-mind at generation time.
Keep the echo short. You are not duplicating the whole instruction — just the single decision the model must not forget.
messages = [
{"role": "user", "content": (
retrieved_docs
+ "\n\n---\nReminder: cite a source URL for every "
"claim. If a fact is absent from the docs above, "
"say so explicitly — do not invent it."
)},
]Trim Verbose Tool Output
A major source of middle-bloat is raw tool output. An API or MCP tool can return hundreds of fields when you need three. Dumping all of it pushes your real signal into the low-attention middle.
Trim tool results to the relevant fields before appending them to the message history. Less noise in the middle means the model spends its attention on what matters.
def trim_order(raw: dict) -> dict:
# keep only what the model needs to reason about
return {
"order_id": raw["order_id"],
"status": raw["status"],
"total": raw["total"],
"refundable": raw["refundable"],
}
# append the trimmed result, not the 200-field blob
messages.append({
"role": "user",
"content": [{
"type": "tool_result",
"tool_use_id": tu_id,
"content": json.dumps(trim_order(raw_result)),
}],
})Keep Case Facts Verbatim
When context grows, teams reach for progressive summarization. Useful — but it makes numbers, percentages, and dates vague. Combined with lost-in-the-middle, a summarized transactional fact sitting mid-context is doubly fragile.
The fix: pull transactional facts into a separate "case facts" block kept verbatim, outside the summary. Don't let an order total or a refund threshold survive only as a paraphrase in the middle of a summary.
case_facts = (
"CASE FACTS (verbatim, do not summarize):\n"
"- order_id: A-4821\n"
"- order_total: $612.40\n"
"- refund_policy_cap: $500.00\n"
"- customer_verified: true"
)
system = case_facts + "\n\n" + rolling_summary/compact Has the Same Risk
In Claude Code, /compact compresses the conversation to free up context. The benefit is room to keep working; the risk is identical to progressive summarization: numbers and dates become vague.
Before compacting a session that hinges on exact values, capture those values somewhere durable — for example in CLAUDE.md via /memory, which persists across sessions — so they survive compaction rather than dissolving into a fuzzy middle summary.
# In a Claude Code session:
/memory # write exact build numbers / API contract to CLAUDE.md
/compact # now safe(r): the hard facts persist outside the summaryOrder Retrieved Documents Deliberately
If you inject N retrieved chunks, their order changes what the model uses. The most relevant evidence should sit at the start or the end of the block — not at chunk number 7 of 14.
Two practical moves:
- Re-rank so the highest-relevance chunks bracket the block.
- Cut the block down. Fewer, sharper chunks beat a long tail of marginal ones diluting attention.
ranked = rerank(query, chunks) # best first
top = ranked[:6] # cut the long tail
# bracket: strongest at start AND end
ordered = [top[0]] + top[2:] + [top[1]]
context = "\n\n".join(c.text for c in ordered)Split Work, Don't Dilute Attention
Lost-in-the-middle is one reason single-pass multi-file review dilutes attention. Cramming ten files into one prompt buries the middle files.
Decompose instead: a per-file local pass, then a separate cross-file integration pass. Each pass has a focused, shorter context where nothing important is stranded in the middle.
# Multi-pass review (avoids single-prompt dilution)
for path in changed_files:
review_local(path) # focused, short context per file
review_cross_file(changed_files) # separate integration passSubagents Need Explicit Context
Multi-agent systems are hub-and-spoke. Subagents do not inherit the coordinator's conversation history — every fact must be passed explicitly in each subagent prompt.
This intersects with lost-in-the-middle: when the coordinator builds that subagent prompt, the critical instructions still belong at the start and end, and verbose blobs still get trimmed. Context isolation is an opportunity to hand each subagent a clean, well-positioned prompt instead of an inherited mess.
Task(
description="Summarize Q3 revenue",
prompt=(
"TASK: extract Q3 net revenue from the report below.\n"
+ report_text +
"\n\nOUTPUT: a single number in USD, plus the source "
"page. If absent, reply 'NOT FOUND'."
),
)A Mental Checklist
Before you ship a long-context prompt, run this checklist:
- Top: most important instruction in
system, stated explicitly. - Bottom: short echo of the single critical constraint at the end.
- Middle: trimmed tool output, re-ranked docs, no raw blobs.
- Facts: exact numbers/dates kept verbatim in a case-facts block, never only in a summary.
- Length: if it's huge, split into focused passes.
Position is design. Treat the middle as low-trust real estate.
Quick Check
Apply the lesson to a concrete architecture decision.
Recap
Key takeaways on the lost-in-the-middle effect:
- Models attend to the start and end more than the middle — treat the middle as low-trust space.
- Put critical instructions in
systemand echo the one constraint that matters at the end. - Trim verbose tool output and re-rank retrieved docs so signal brackets the context.
- Keep exact numbers and dates verbatim in a case-facts block; summarization and
/compactmake them vague. - For large work, split into focused passes rather than one diluted prompt.
- Position mitigates the symptom — but for rules with financial, legal, or safety stakes, enforce with hooks, not placement.
Domande Frequenti
La lezione «Effetto «lost in the middle»» è gratuita?
Sì — il testo completo di «Effetto «lost in the middle»» è gratuito qui sul web. Per esercitarvi in modo interattivo (un editor di codice integrato e un tutor IA 24/7) e sbloccare il resto del corso Claude Architect, passa a CoddyKit PRO. Il corso Claude Architect include 4 lezioni in totale.
Cosa imparerò in «Effetto «lost in the middle»»?
I modelli leggono l’inizio e la fine in modo più affidabile rispetto alla parte centrale. Eserciti Claude Architect con codice pratico che esegui direttamente nel browser, e un tutor IA 24/7 risponde alle tue domande mentre lavori sulla lezione.
Ho bisogno di esperienza per iniziare Claude Architect?
Non è richiesta alcuna esperienza precedente. Claude Architect su CoddyKit è strutturato per principianti e studenti avanzati, quindi puoi iniziare da qui o dall'inizio e procedere al tuo ritmo. Questa è la lezione 3 di 4.
Quanto tempo richiede la lezione «Effetto «lost in the middle»»?
La maggior parte delle lezioni CoddyKit richiede circa 5–10 minuti. Ogni lezione è breve e interattiva, quindi fai progressi costanti e riprendi esattamente da dove hai lasciato su web e app.
Posso scrivere ed eseguire codice in questa lezione Claude Architect?
Sì. Ogni lezione Claude Architect include un editor di codice integrato, quindi scrivi ed esegui codice reale direttamente nel tuo browser e ricevi feedback istantaneo dall'IA — nessuna configurazione locale necessaria.
Tutte le lezioni di questo corso
- È necessaria l’intera cronologia
- Rischi del riepilogo progressivo
- Effetto «lost in the middle»
- Blocchi dei fatti del caso e riduzione dell’output