19 / papers

Multi-Agent Systems

What emerges when models talk to each other instead of to us. 19 open-access papers by Pranay Mahendrakar, each with a permanent DOI.

4 Oct 2026 article

The Evaluator Is the Bottleneck: What a Self-Modifying System's Acceptance Test Must Satisfy Before Its Verdict Can Be Trusted, Why Co-Evolving Evaluators Shrink the External Anchor Rather Than Remove It, and How Moving the Anchor Outside Moves the Attack Surface With It

Self-modifying AI systems now decide which changes to their own code to keep by running an acceptance test and retaining whatever scores higher. Published coding-agent loops report large benchmark gains this way, and their safety case…

DOI

3 Oct 2026 article

Whose Goals? Autotelic Agents Generate Goals Within Spaces They Are Given: The Goal Space and the Referee Stay Outside the Agent, and a Foundation Model in the Loop Relocates Them Rather Than Removing Them

Autotelic agents are described as learning to represent, generate, select and solve their own goals, and a new generation of systems built on foundation models is reported to do so without hand-coded goal representations, without human…

DOI

2 Oct 2026 article

Pooling Experience Spends Independence Twice: Shared Memory in LLM Agent Teams Correlates Errors and Opens a Common-Mode Channel, Why Little of the First Was Left to Spend, and the Order, Gate and Readout That May Decide Whether Pooling Pays

Teams of language-model agents increasingly read and write a common memory: a message pool, a blackboard, a bank of distilled experience, a knowledge base served to several frameworks at once. The case for pooling is efficiency, since no…

DOI

1 Oct 2026 article

Perspectives Without Independence: Multi-Agent and Multi-Persona Reasoning Under Compute-Normalised Comparison, Why the Gains That Survive Are Not the Ones Diversity Predicts, and the Controls That Would Tell Them Apart

Multi-agent debate, multi-persona prompting and related schemes are usually justified by diversity of viewpoint: several perspectives err differently, so their combination is more reliable than any one of them. That justification is a…

DOI

27 Sep 2026 article

Consensus Too Soon, or Agreement From the Start? Shared Prior, Social Coupling and Pool Coverage in Decentralised LLM Collectives, and Why Prompted Diversity and Model Heterogeneity Act on Different Terms

Groups of language-model agents that exchange answers and settle on a common one are now a standard way to build decentralised decision systems. A common worry is that they agree too soon: agents copy each other, diversity collapses, and…

DOI

26 Sep 2026 article

Compressed Once, Read Many Times: Why Prompt-Compression Results Do Not Transfer to Agent Memory, What the Query-Agnostic Line Already Showed, and the Two Properties of a Memory Write No Protocol Yet Scores

Language-model agents that remember across sessions compress what they store: they summarise dialogue, extract facts, or evict cache entries, and then answer later questions from what is left. The compression ratios used to justify these…

DOI

25 Sep 2026 article

Checked at Every Step Is Not Checked as a Whole: Two Senses of Plan-Level Safety for LLM Agents, and Why Decomposition Attacks Exploit the Gap Between Them

Most deployed safety mechanisms for LLM agents judge one unit at a time: a single tool call, or a single (observation, action) pair (Choi et al., 2026). A separate literature asks whether that is the right unit to judge at all. Jones…

DOI

22 Sep 2026 article

Three Things Called Budget Awareness: Observability, Forecasting and Allocation in LLM Agents, Why Every Published Allocation Gain Is Keyed to a Signal Measured After the Fact, and the Run-to-Run Variance No Forecast Is Scored Against

Two 2024-2026 literatures make claims about resource use in language-model agents that look incompatible. One reports that allocating test-time compute according to problem difficulty beats spending it uniformly, by margins up to a factor…

DOI

21 Sep 2026 article

Scored Before the Question Exists: What a Write-Time Importance Value in LLM Agent Memory Predicts, Why an Additive Term Is Not a Prior, and the Ablation the Canonical Architecture Did Not Run

An agent that stores what happens to it must decide what is worth storing and, later, what is worth reading back. The most-copied mechanism for the first decision is a scalar written at storage time: a language model is shown a record and…

DOI

17 Sep 2026 article

Not Acting Is Not One Decision: Two Abstention Triggers in Tool-Using Agents, Why the Calibration and Permission Readings Cover Different Ones, and the Cost Term Neither Supplies

An agent that declines to send the email has made a decision that looks like the decision a language model makes when it declines to answer a question, and the resemblance has organised the 2026 literature into two camps. One reads…

DOI

16 Sep 2026 article

A Lesson Is an Untested Counterfactual: The Three Claims Inside a Stored Agent Failure Memory, Which of Them Anything Checks, and Why Every Published Retention Rule Keys on a Proxy

An agent that fails a task and writes down what it should have done instead has produced a specific kind of object: a sentence asserting that a named step caused the failure and that a named alternative would not have. This paper takes…

DOI

11 Sep 2026 article

Consolidation Without Weights: What the Complementary Learning Systems Analogy Licenses in LLM Agent Memory, and Why the Systems That Borrow Its Name Do Not Inherit Its Guarantee

Memory systems for language-model agents almost all contain a step called consolidation, and almost all of them cite, or gesture at, the complementary learning systems account of hippocampus and neocortex when they name it. In that account…

DOI

10 Sep 2026 article

Attribution Is Scored on a Finished Trace: What Failure-Attribution Benchmarks Measure, What Cascade Containment Would Need, and Why Neither Result Bounds the Other

A family of recent proposals aims to stop errors from spreading through LLM-based multi-agent systems: genealogy-graph governance over message dependencies, cross-channel causal monitoring, propagation-aware remediation of contaminated…

DOI

6 Sep 2026 article

Delete Names Five Operations: Erasure Semantics for Derived Agent Memory, and Why a State-Level Definition Cannot Be the Auditable One

Persistent memory has become a standard component of language-model agents, and with it a standard assumption: that where a fact lives in a retrievable store rather than in weights, erasing it is a solved problem, because a store supports…

DOI

3 Sep 2026 article

A Safety Memory Is a Declassification Channel: Why Origin Binding, Taint Tracking and Memory Isolation Do Not Secure the Records an Agent Writes About Attacks

Two lines of work on language-model agents have converged on the same component from opposite directions. One proposes that an agent keep a persistent safety memory, so that an attack seen once, a request refused once or a source found…

DOI

31 Aug 2026 article

Newer Is Not Truer: Three Presuppositions Behind the Recency Rule in Long-Term Agent Memory, and Why Retaining Both Records Relocates the Decision Rather Than Making It

An agent with persistent memory writes its own records, and two of them can disagree. Across the published systems examined here the resolution runs in one direction: the newer record supersedes the older. This paper argues that the…

DOI

28 Aug 2026 article

Ordering Is Not Resolution: What the Instruction Hierarchy Defines, What It Leaves Undefined, and the Same-Tier Conflicts Agent Benchmarks Do Not Separate

When a language-model agent receives instructions that conflict, the dominant remedy is a privilege ordering over sources: system above developer, developer above user, user above tool output. This paper argues that the ordering paradigm…

DOI

21 Aug 2026 article

A Second Model Is Not a Second Opinion: Peer Verification, Correlated Failure, and the Independence Variable Nobody Reports

When a language model cannot reliably detect its own errors, the standard remedy is to have a second model check the first. That substitution is now the default across agent benchmarks, reward pipelines and safety evaluations, and it rests…

DOI

28 Apr 2026 article

Emergent Covert Signaling in Multi-Agent LLM Negotiation: A Conceptual Framework and Experimental Protocol

When multiple large-language-model agents negotiate, communicate, or compete, do they spontaneously develop covert signalling — channels of communication that human observers cannot decode? Recent work has established that such behaviour…

DOI

About this research line

What has Pranay Mahendrakar published on Multi-Agent Systems?

Pranay Mahendrakar has published 19 open-access papers on Multi-Agent Systems: "The Evaluator Is the Bottleneck", "Whose Goals? Autotelic Agents Generate Goals Within Spaces They Are Given", "Pooling Experience Spends Independence Twice", "Perspectives Without Independence", "Consensus Too Soon, or Agreement From the Start? Shared Prior, Social Coupling and Pool Coverage in Decentralised LLM Collectives, and Why Prompted Diversity and Model Heterogeneity Act on Different Terms", "Compressed Once, Read Many Times", "Checked at Every Step Is Not Checked as a Whole", "Three Things Called Budget Awareness", "Scored Before the Question Exists", "Not Acting Is Not One Decision", "A Lesson Is an Untested Counterfactual", "Consolidation Without Weights", "Attribution Is Scored on a Finished Trace", "Delete Names Five Operations", "A Safety Memory Is a Declassification Channel", "Newer Is Not Truer", "Ordering Is Not Resolution", "A Second Model Is Not a Second Opinion" and "Emergent Covert Signaling in Multi-Agent LLM Negotiation". Each is deposited on Zenodo with a permanent DOI and is free to read.

← All research topics